One page replaces Research, Conditions, and About
Learn

How we grade. What we won't grade. Why. Six tabs replace the current nav structure's four separate sections.

The four rules — no exceptions

1

Human RCTs only

Mechanistic and animal data is labeled separately and never enters a grade. A supplement with 500 mouse studies and 0 human trials is Tier D. Period. This rule is the hardest to hold because the longevity field produces compelling animal data constantly. We hold it anyway.

2

Absolute risk + NNT

We show what the trial actually moved — not the relative-risk headline. "Reduces heart attack risk by 25%" is useless without the baseline rate. We report the number needed to treat (NNT) to prevent one event, and the absolute risk reduction. This is the standard Cochrane uses.

3

Dose and form must match

A grade applies only to the dose and chemical form that was studied. "Magnesium" is not one compound — glycinate, oxide, and threonate have different pharmacokinetics and different evidence bases. A grade for one form does not transfer to another. We flag this on every product card.

4

No affiliate influence

Affiliate relationships generate revenue for EBL but are disclosed and never influence ratings or rankings. A Tier A supplement that we don't earn commission on remains Tier A. A Tier C supplement that we do earn commission on remains Tier C. This rule is publicly verifiable — the grades are documented before any affiliate agreements are signed.

How grades are assigned

Step 1 — Literature search
PubMed search for human RCTs within the past 15 years. Earlier foundational trials included if they remain the primary evidence base. No preprint servers — only peer-reviewed publications.
Step 2 — Endpoint classification
Trials are classified by endpoint: hard outcome (mortality, CVD event, fracture), biomarker (HbA1c, LDL-C, bone density), mechanistic (gene expression, in vitro). Only hard outcome and biomarker trials count toward a grade.
Step 3 — Replication check
Tier A requires replication across at least 2 independent research groups. A single landmark trial can support Grade B but not Grade A. Cochrane reviews count as replicated synthesis.
Step 4 — Conflict of interest screen
Industry-funded trials are weighted lower than independently funded trials. Consistent results across industry and independent funding → grade stands. Industry-funded-only results → grade reduced by one tier.
Step 5 — Annual review
All grades are reviewed on a 12-month cycle. New RCTs are evaluated within 30 days of publication in high-impact journals. Grade changes are logged in the Evidence Log with date and rationale.

Four tiers, not a continuous spectrum. A supplement either has hard human outcome data or it doesn't. The tier tells you what kind of evidence exists — the grade within each tier tells you how consistent it is.

A
Landmark
evidence
Multiple high-quality RCTs in humans. Hard endpoints (mortality, CVD events, fractures). Effect sizes replicated across independent research groups. Decades of safety data. Dose and form clearly defined. Examples: Omega-3 EPA/DHA (REDUCE-IT), Creatine (500+ RCTs), Magnesium (34-RCT meta-analysis).
B
Strong
biomarker
signal
Human RCTs with biomarker endpoints (HbA1c, LDL-C, bone density, VO₂max, cognitive scores). Consistent signal across multiple trials. No hard outcome trial yet — or hard outcome trial in progress. The largest category on EBL: most well-researched supplements live here. Examples: Berberine, CoQ10, NMN, Ashwagandha.
C
Promising
early signal
Small human trials, mixed results, or one positive RCT without replication. Often the "exciting animal data didn't translate cleanly" category. Not a reason to dismiss — a reason to watch carefully and retest personally. Examples: Resveratrol, Quercetin, Spermidine, Lion's Mane.
D
Mechanistic
only
No human RCT data. Animal studies, in vitro, expert opinion, or mechanistic hypothesis only. ~90% of preclinical findings fail to replicate in humans — this is the baseline expectation at Tier D, not an anomaly. We include Tier D with clear labeling because mechanisms matter; they just don't substitute for human data.
How EBL grades compare to Examine.com: Examine grades effectiveness at the primary stated endpoint. EBL grades specifically against longevity endpoints (cardiovascular events, cancer mortality, biological age markers). The same supplement can have different grades because we're asking different questions — not because one assessment is wrong. See full concordance table →
No. EBL has no equity stake, employment relationship, or editorial control agreement with any supplement manufacturer or brand. We use affiliate links (disclosed on every page) which generate revenue when you purchase through our links — but affiliate relationships are never a factor in grading. A supplement receives its tier based solely on published human trial data.
Because the supplement industry has a significant marketing-to-evidence gap. Resveratrol is heavily marketed based on compelling animal data — the human translation has been disappointing. Vitamin C megadosing has an enormous cultural footprint — the RCT data for longevity endpoints is weak. Our grades follow the evidence, not the marketing spend or the cultural momentum.
Absolutely. Grades reflect population-level evidence, not individual responses. A Grade A supplement can interact with your specific medications, be contraindicated for your health conditions, or simply not produce the expected biomarker response in you. This is why EBL emphasizes testing before and after — to confirm that the population-level finding is replicating in you specifically.
Annual full review for all grades. New RCTs published in high-impact journals (NEJM, JAMA, Lancet, BMJ, Nature Medicine, Cell Metabolism) are reviewed within 30 days of publication and can trigger grade changes outside the annual cycle. All grade changes are logged in the Evidence Log with date, reason, and the study that triggered the change.
All safety tools and evidence grades are free and always will be. Pro adds: the Marker Trend Viewer (longitudinal biomarker tracking), Evidence Log (structured n=1 experiment tracking with export), Stack Comparison (compare two stacks head-to-head), and early access to new grades before public publication. Pro is $9/month — less than one month of most Grade A supplements.
No. EBL is educational content only. Nothing here constitutes medical advice, and nothing should be started, stopped, or changed without discussion with your physician — especially medications. The Physician Discussion Sheet tool exists specifically to help you bring this information to a clinical conversation rather than act on it alone.
The founder is shomer Shabbat. The Kosher Longevity site (kosherlongevity.com) is the dedicated resource — EBL's kosher filter surfaces it within the main platform. Halachic nuances (capsule shell type, fermentation substrates, Pesach kitniyot status) matter and are not typically addressed by supplement manufacturers.

A public record of grade changes — what changed, why, and what study triggered it. This is the transparency layer that most supplement sites lack. Every update is logged here.

May 2026
NMN — maintained at Grade B
Reviewed 3 new RCTs. Yoshino 2021 remains the strongest human trial. New 2024 trial in older adults showed NAD+ elevation but no significant functional improvement at 12 weeks. Grade B maintained — promising but needs replication of the functional endpoint.
Apr 2026
GlyNAC — upgraded Grade C → Grade B
Kumar 2023 (JNHA) published and reviewed. Randomized, double-blind, 24-week trial: improved strength, cognition, and mitochondrial function. Sufficient for B. Replication needed for Grade A consideration.
Mar 2026
Vitamin D3 — maintained at Grade A (longevity-specific endpoints)
VITAL extended follow-up published. Cancer mortality reduction at 2,000 IU confirmed at 5 years (HR 0.71). Bone fracture endpoint null in primary prevention. EBL grade reflects the cancer mortality endpoint specifically. Examine.com maintains Grade B reflecting mixed endpoint picture — both assessments are defensible.
Feb 2026
Resveratrol — maintained at Grade C
New bioavailability review confirms the fundamental pharmacokinetic limitation at oral doses. Transresveratrol absorption 1% in most adults. No new human outcome trial. Grade C appropriate — mechanism compelling, human translation poor.
Jan 2026
Berberine interaction flag updated
New case series: 3 documented hypoglycemia events when berberine combined with metformin at full dose without physician oversight. Interaction flag upgraded from "note with physician" to "physician oversight required." Affected the Medication Reality Check tool output and the Interaction Screener.
EBL
The Founder — credentials withheld during California DC board review
Longevity Researcher & Clinical Nutritionist · A4M Diplomate · Shomer Shabbat

Founder of Evidence Based Longevity and Kosher Longevity. Decades of clinical nutrition and longevity practice. A4M (American Academy of Anti-Aging Medicine) Diplomate. Currently completing California chiropractic licensure — full credentials will be listed upon license issuance. Independent — no financial stake in any product, brand, or laboratory on this site.

Why EBL exists
After decades watching patients spend money on supplements that had no meaningful human evidence — while ignoring drug-nutrient interactions that were quietly depleting their B12, magnesium, and CoQ10 — the need for a medication-first, evidence-first resource was obvious. Most longevity platforms exist to sell products. EBL exists to grade them.
What makes EBL different
The "medication first before supplementing" approach is EBL's core differentiator. Most longevity sites assume a healthy, unmedicated adult. Most supplement users are not that person. Checking medication interactions and nutrient depletions before building a supplement stack is clinical thinking — not a feature, a prerequisite.
Conflicts of interest
EBL uses affiliate links (disclosed on every page). The Fullscript dispensary earns a commission when you purchase through EBL. No equity, employment, or editorial arrangement exists with any supplement brand. All affiliate arrangements postdate grade assignments and cannot change them retroactively.

The intersection of kashrut and clinical longevity nutrition is genuinely complex. Most supplement manufacturers don't address halachic nuances. EBL and Kosher Longevity do. This is a summary — the full resource is at kosherlongevity.com.

Capsule shells
Gelatin capsules — bovine or porcine gelatin is not kosher without specific certification. Fish gelatin may be permitted — consult your posek.

Vegetable/plant capsules — HPMC or pullulan capsules are generally pareve and kosher without concern.

Softgels — fish gelatin softgels are acceptable for most opinions. Bovine softgels require glatt certification.
Fermentation substrates
NMN and NR — some brands use yeast fermentation. Consult your posek on yashan/chadash status if applicable.

Vitamin D3 — lanolin-derived (from sheep wool) vs. lichen-derived (pareve). Lanolin D3 is generally permitted; lichen D3 is strictly pareve.
Pesach considerations
Kitniyot — many supplements use corn, soy, or legume-derived excipients. Ashkenazim avoiding kitniyot should check each product.

Chametz — wheat germ-based supplements (spermidine) require kashrut lePesach certification.
Verified certifications
OU, OK, Star-K, Kof-K — major certification bodies. Look for the hechsher on the supplement label or the brand's website.

OU Fish — indicates fish-derived gelatin in softgels.

OU Pareve — no meat or dairy. Safe for all meals.
Full Kosher Longevity resource →