Methodology

This page documents the citation standard used in the Nutri Tailor Health Reference Library. It exists because every claim in a clinical reference work is only as good as the rules that put it there. We make ours public.

The standard aligns with the British Association for Nutrition and Lifestyle Medicine (BANT) approach, layered with the Complementary and Natural Healthcare Council (CNHC) definition of evidence-based practice. Five sourcing layers, an evidence hierarchy, and explicit exclusions.

Layer 1 — BANT-curated databases

Highest tier. If a study is included in Nutrition Evidence Database (NED), the BANT-curated scientific database, it has passed an editorial gate that requires high-quality human studies — RCTs, systematic reviews, meta-analyses, and cohort studies. NED explicitly excludes animal and cell studies. For supplement, herb, vitamin, mineral, and drug-supplement interaction claims, we cross-check against NatMed Pro, the licensed BANT member resource.

Layer 2 — Evidence hierarchy

Level	Evidence type	Use in the library
1	Systematic reviews, meta-analyses of RCTs, Cochrane Reviews	Cited by preference where they exist.
2	Individual randomised controlled trials in humans	Standard.
3	Prospective cohort studies	Acceptable where higher evidence isn’t available.
4	Cross-sectional, case-control, narrative reviews	Used cautiously. Flagged in prose (“limited evidence”, “preliminary”).
5	Mechanistic / animal / in vitro studies	Not citable directly for human clinical claims. Mechanism context drawn from Level 1–3 human studies or institutional reviews.
6	Expert opinion, anecdote, single case reports	Never citable as standalone evidence.

Layer 3 — Statutory and authoritative bodies

UK statutory and quasi-governmental:

NICE (National Institute for Health and Care Excellence) — clinical guidelines
NHS (gov.uk and nhs.uk)
SACN (Scientific Advisory Committee on Nutrition)
BNF (British National Formulary)
MHRA (Medicines and Healthcare products Regulatory Agency)
FSA (Food Standards Agency)

International:

WHO (World Health Organization)
EFSA (European Food Safety Authority)
NIH ODS (US National Institutes of Health, Office of Dietary Supplements)
Endocrine Society, Cochrane

The library aligns with statutory guidance where it exists. We do not contradict NICE, NHS, or SACN positions without strong, explicit caveats.

Layer 4 — Bibliographic indexes

Studies cited in the library must be properly indexed in at least one of:

PubMed/MEDLINE (NIH/NLM)
Cochrane Library
EMBASE
CINAHL
Web of Science

Each peer-reviewed source carries either a PMID or DOI on the entry page. Both are clickable links to the canonical record.

Layer 5 — Peer-reviewed journals

Examples of journals routinely cited: BMJ, British Journal of General Practice, The Lancet, Lancet Diabetes & Endocrinology, JAMA, New England Journal of Medicine, American Journal of Clinical Nutrition, Nutrients, Journal of Clinical Endocrinology and Metabolism, Journal of Nutritional Biochemistry, Haematologica.

Any reputable peer-reviewed nutrition, medical, or public-health journal indexed on PubMed is acceptable. The test is study quality (Layer 2 hierarchy), not journal name alone.

Explicitly not acceptable

Manufacturer claims, supplement company white papers, marketing material
Unrefereed blogs, podcasts (other than curated content like BMJ Talks Medicine)
Social media as a primary source
Anecdotal reports or single case reports as standalone evidence
Animal-only or in vitro-only studies cited as evidence for human clinical claims
Predatory or non-peer-reviewed journals
Aggregator sites used as primary sources (Examine.com, Healthline, WebMD, Wikipedia, Mayo Clinic articles) — these are research tools, not citable sources

Industry funding — distinguishing two cases

The library treats industry funding at the individual study level, not the platform level.

Not acceptable: individual studies with undisclosed industry funding, industry-influenced study design, or author conflict of interest that compromises independence.
Acceptable: ethical sponsorship of curation infrastructure (for example, NED’s openly-disclosed industry partners supporting the platform without compromising the independent Editorial Board and Expert Reviewer panel).

The test is on the cited paper itself: does it carry an undisclosed conflict of interest, or industry-influenced design?

CNHC overlay — evidence-based practice

The library follows the CNHC Code of Conduct, Ethics and Performance definition of evidence-based practice (Sackett et al. 2000): research evidence + clinical expertise + patient values, weighted equally.

This shapes the voice. Entries are suggestive, not prescriptive. Readers are pointed to the evidence and to qualified professionals. The library does not tell anyone what to take.

UK supplement-availability rules

The following substances are not authorised for sale as food supplements in the UK and the library will not recommend them:

DIM (diindolylmethane)
CBD (cannabidiol, as a food supplement)
GABA (gamma-aminobutyric acid, as a supplement)
Melatonin (prescription-only in the UK)

These substances may appear in mechanism, biology, or interaction-warning context — for example, an entry on sleep may reference endogenous melatonin synthesis, or warn about supplement interactions for readers already taking melatonin under prescription.

Specialist-CPD conditions — defer to specialist

Entries touching the following conditions defer to specialist clinicians and do not imply primary-care recommendations: confirmed or suspected cancer, type 1 diabetes, epilepsy, post-bariatric/gastric-band, and colostomy. These conditions require specialist post-qualification training under BANT and CNHC scope-of-practice rules.

Confidence grading

Every entry’s underlying evidence quality is graded:

Strong — multiple Level 1 systematic reviews or large RCTs supporting the claim
Moderate — one or more Level 2 RCTs, or consistent Level 3 cohort evidence
Limited — Level 3 or Level 4 evidence only, or contested findings

The grade appears on every entry. Where the underlying evidence is weak, the prose says so explicitly.

Corrections and updates

If you find an error, a stale citation, or a guideline that has changed, please write to info@nutritailor.co.uk. Every entry has a per-entry change log visible to readers. Substantive corrections are logged with date and reason.

How to cite

Nutri Tailor Health Reference Library, edited by Henry Bond. Available at https://nutritailor.co.uk/apps/learn. Last-reviewed dates appear on each entry.