Data Baselines
Last updated:
1. Purpose
THE PROTOCOL rates longevity interventions — supplements, protocols, devices, and practices — using a structured, evidence-based framework. This page explains how we gather data, evaluate evidence, assess risk, and arrive at ratings. Transparency in methodology is non-negotiable.
2. The Evidence × Risk × Cost Framework
Every protocol rating is built on three independent axes:
2.1 Evidence Quality
We grade evidence on a five-tier hierarchy. Higher tiers carry more weight in rating calculations:
| Tier | Source Type | Weight |
|---|---|---|
| T1 | Meta-analyses, systematic reviews of RCTs | Highest |
| T2 | Randomized controlled trials (RCTs) | High |
| T3 | Cohort studies, observational data | Moderate |
| T4 | Case studies, animal models, in vitro | Low |
| T5 | Expert opinion, traditional use, mechanistic plausibility | Minimal |
We note study size, duration, population demographics, and funding sources. Industry-funded studies are flagged but not automatically discounted — we assess methodology independently.
2.2 Risk Profile
Risk assessment considers:
- Known side effects and their severity/frequency
- Drug interactions and contraindications
- Long-term safety data availability (or lack thereof)
- Regulatory status (approved, generally recognized as safe, unregulated)
- Dose-response curve — therapeutic window vs. toxicity threshold
2.3 Cost Efficiency
We evaluate cost per unit of expected benefit. A $5/month supplement with solid T2 evidence outranks a $200/month protocol backed only by mechanistic reasoning. Cost analysis includes:
- Direct cost (purchase price at typical dosing)
- Required testing/monitoring costs
- Opportunity cost vs. alternative interventions
- Availability and supply chain reliability
3. Source Selection
We pull data from:
- PubMed / MEDLINE for peer-reviewed research
- Cochrane Library for systematic reviews
- ClinicalTrials.gov for ongoing and completed trials
- FDA, EMA, and TGA regulatory databases
- Examine.com and similar evidence aggregators (as secondary references)
- Preprint servers (bioRxiv, medRxiv) — flagged as unreviewed
We do not use manufacturer claims, influencer testimonials, or single-source promotional material as evidence inputs.
4. Rating Scale
Protocols receive a composite rating that synthesizes evidence quality, risk profile, and cost efficiency. The scale is intentionally simple:
| Rating | Interpretation |
|---|---|
| STRONG | Robust evidence, favorable risk profile, reasonable cost. High confidence recommendation. |
| MODERATE | Meaningful evidence with some gaps. Benefits likely outweigh risks for most people. |
| EMERGING | Promising early data. Worth watching. Not enough evidence for a confident recommendation. |
| INSUFFICIENT | Claims exceed evidence. May have potential but data is too thin to rate meaningfully. |
| AVOID | Risk outweighs demonstrated benefit. Evidence contradicts common claims, or safety concerns are material. |
5. Update Frequency
Ratings are reviewed when significant new evidence is published — typically when a new meta-analysis, large RCT, or safety signal emerges. We do not update on a fixed schedule. Each rating page shows its last review date and a changelog of material revisions.
6. Conflict of Interest Policy
THE PROTOCOL may use affiliate links to supplement retailers. Affiliate relationships never influence ratings. A protocol rated "AVOID" will carry that rating regardless of whether we have an affiliate relationship with its manufacturer. We disclose all material conflicts at the point of recommendation.
7. Limitations
Our methodology has inherent limitations:
- We rely on published research, which has its own publication bias
- Individual responses to interventions vary based on genetics, health status, and context
- Cost data reflects typical US/EU pricing and may not apply globally
- Evidence quality in longevity science is generally lower than in acute medicine — long-term human RCTs are rare and expensive
- We apply editorial judgment in synthesizing evidence — reasonable people may weigh the same data differently
8. Methodology Feedback
We welcome scrutiny of our methodology. If you believe a rating misrepresents the evidence, cite the specific studies and we will review. Contact information is provided on the Site.