Methodology
How we grade the evidence.
NutriDex is a free, independent health-literacy project, updated weekly. For every substance we summarise the human research, assign a transparent evidence tier, and link the studies behind it — so you can see not just what a supplement claims, but how strong the evidence for it actually is. Here's exactly how that works.
Why NutriDex exists
A note on what drove this project.
NutriDex was created by Dr Daryl Peh out of a deep — admittedly obsessive — commitment to evidence-based medicine. He is the sort of clinician who reads the original papers, checks the sample sizes and study designs, and cares far more about the weight and quality of the evidence than the confidence of the marketing around a product.
The motivation is simple: supplement and "wellness" claims routinely outrun the data, and the public rarely gets to see how thin — or how solid — the underlying research actually is. NutriDex is an attempt to fix that — a meticulous, plain-English audit of what the human studies genuinely show for each substance, graded honestly and linked to the source, so anyone can apply the same scrutiny to their own choices that a careful clinician would demand for themselves.
The six evidence tiers
Each entry gets one tier, reflecting the weight and consistency of human research as of early 2026.
Inside each entry you'll also see an "evidence strength" meter (one to five bars). It's a quick visual of the same judgement — more bars mean a heavier, more consistent body of human evidence behind the headline claims.
"Studies" vs "Verified"
Two different numbers you'll see on cards and in each dossier.
★ Studies
The total number of sources we cite for that entry. Entries with 10 or more carry a "★ N studies" badge, marking the most heavily-researched substances.
✓ Verified
How many of those citations link straight to the original source — a PubMed record, a DOI, or an official body (NIH, FDA, Cochrane, WHO, WADA). You can click through and read it yourself.
So studies is "how much research is behind this," and verified is "how much of it you can open and check."
How we choose which studies to show
Well-researched substances can have a dozen-plus citations — so we lead with the strongest.
Each entry's detail view shows the six highest-quality citations first, with the rest one tap away under "Show more." Citations are ranked by study type and recency, in roughly this order:
- Meta-analyses, systematic reviews & Cochrane reviews — syntheses of many studies.
- Authoritative guidelines & regulators — NIH/NCCIH/ODS, FDA, EFSA, WHO, WADA, major society position stands.
- Randomized controlled trials (RCTs) — especially large, recent, or landmark ones.
- Reviews, observational & cohort studies, then mechanism, preclinical and case reports.
Recent work (2021–2026) is weighted higher, so the newest robust evidence rises to the top. Findings are written to state the direction and rough size of the effect and the population studied — not just "a study showed…".
Where the evidence comes from
Every citation links to a reputable, checkable source.
We also flag honest caveats inside summaries — for example when a positive trial was industry-funded, when products vary wildly in content, or when an effect is real but small.
Independence & limits
NutriDex is editorially independent and free to read. Its content is not for sale: it recommends no specific commercial products, carries no sponsored entries, and uses no affiliate links. The database is updated weekly to stay current. It is curated and reviewed by a licensed medical practitioner holding a Master of Medicine (MMed), practising in Singapore.