Evidence Grading Methodology
How we evaluate and grade probiotic and prebiotic claims - and why most of what you read online gets this wrong.
Why Evidence Grading Matters Here
The probiotic and prebiotic space has a specific and well-documented credibility problem: supplement brands overclaim, and medical institutions under-communicate. Supplement brand content says "probiotics boost immunity, improve mood, burn fat, and cure IBS." Mayo Clinic says "evidence is inconclusive." Both statements fail the reader, for opposite reasons.
The truth is more specific and more interesting: some strains have strong evidence for specific conditions. L. rhamnosus GG has Cochrane-level evidence for reducing acute diarrhoea duration in children by one day. Psyllium husk has EFSA-recognised evidence for LDL cholesterol reduction and NICE-endorsed guidance for constipation. S. boulardii reduces antibiotic-associated diarrhoea risk by 54% (Cochrane 2015). These are not vague wellness claims - they are specific, quantified, replicated effects.
Our methodology is designed to communicate this specificity honestly, including when the evidence does NOT support a popular claim.
The Three Evidence Tiers
Reliably established, replicable effect
A claim receives Strong Evidence when it meets ALL of the following:
- Multiple well-designed, adequately powered randomised controlled trials (RCTs) with consistent findings
- OR at least one high-quality systematic review or Cochrane meta-analysis confirming the effect
- OR formal regulatory recognition: EFSA health claim approval, NICE clinical guideline recommendation, or ISAPP consensus endorsement
- Evidence is strain-specific (not just genus-level) and dose-matched to clinical trials
- Independent replication (not only industry-funded trials)
Examples of Strong Evidence claims: S. boulardii for antibiotic-associated diarrhoea prevention; L. rhamnosus GG for acute paediatric diarrhoea; psyllium for LDL cholesterol reduction; beta-glucan for LDL cholesterol reduction; B. lactis BB-12 for constipation; psyllium for stool regularity in functional constipation.
Promising but not yet consistently replicated
A claim receives Emerging Evidence when:
- At least one well-designed RCT shows a positive effect, but independent replication is limited
- OR consistent findings in preliminary human studies (pilot RCTs, open-label trials) not yet confirmed by a systematic review
- OR strong mechanistic evidence with early human confirmation (not purely animal data)
- Industry-funded trials are eligible for Emerging, not Strong, unless independently replicated
- The effect is strain-specific but the evidence base doesn't yet meet the Strong tier threshold
Examples of Emerging Evidence claims: B. infantis 35624 for IBS symptom severity; L. plantarum 299v for IBS-D; B. longum NCC3001 for IBS-associated depression scores; acacia fibre for IBS microbiome support; synbiotics for constipation improvement over probiotic alone.
Anecdotal, marketing claim, or single low-quality study
A claim receives Limited Evidence when:
- Evidence is limited to animal studies or mechanistic laboratory data without confirmed human RCT
- OR single small, industry-funded study with no independent replication and potential for significant bias
- OR the claim extrapolates from genus-level findings to a specific product ("all Lactobacillus strains help IBS")
- OR the claim is based on consumer testimonials, brand-sponsored content, or anecdote
- OR the claim is scientifically plausible but completely unsubstantiated by human clinical trials
Examples of Limited Evidence claims: "Probiotics for weight loss" (very weak human evidence); "Kombucha for gut health" (SCOBY is not Lactobacillus/Bifidobacterium, human trial evidence is minimal); "Probiotic for skin health" (emerging mechanistic data, no strong human RCTs); "Apple cider vinegar as a probiotic" (Acetobacter is not a recognised probiotic organism).
Sources We Prioritise
In descending order of weight:
- Cochrane systematic reviews: The gold standard for intervention evidence synthesis. We cite these when available.
- ISAPP consensus papers: The International Scientific Association for Probiotics and Prebiotics publishes expert consensus definitions and evidence reviews. These are the authoritative reference documents for our field.
- NICE clinical guidelines: UK National Institute for Health and Care Excellence evidence-based clinical guidance. Particularly relevant for IBS, constipation, and antibiotic prescribing.
- EFSA health claim assessments: European Food Safety Authority scientific opinions on health claims. An EFSA-approved claim is a Strong Evidence anchor.
- Pre-registered RCTs in peer-reviewed journals: PubMed-indexed trials, with preference for independent (non-industry-funded) trials.
- Network meta-analyses: Where multiple RCTs compare different interventions, network meta-analyses provide relative rankings. We rely on Ford et al. (Am J Gastroenterol 2024) for IBS strain ranking.
- Industry-funded RCTs: Considered but assigned Emerging at best unless independently replicated. Conflicts of interest are noted where relevant.
- Mayo Clinic, Harvard Health, Cleveland Clinic: Referenced for clinical context and consumer-facing framing, but not as primary evidence.
We do not cite: press releases, brand white papers, influencer content, consumer testimonials, or mechanistic studies as primary evidence for clinical claims.
Strain Specificity Principle
This is the most important and most frequently violated principle in probiotic communication. Evidence for one strain does NOT transfer to another strain of the same species, much less the same genus. L. rhamnosus GG (ATCC 53103) has Cochrane-level evidence for acute diarrhoea. Another L. rhamnosus strain from a different manufacturer with no strain designation has no established evidence and cannot borrow LGG's credibility.
On this site, we cite the specific strain designation whenever available (e.g. L. rhamnosus GG, B. infantis 35624, S. boulardii CNCM I-745) and we note when a product does not disclose its strain designation. Lack of strain disclosure is itself a quality signal.
Conflict of Interest Transparency
This site uses affiliate links to supplement brands. Where affiliate links exist, they are disclosed on each page. Our affiliate relationships do NOT influence evidence tier assignments. If a product's evidence is Limited, we say so plainly regardless of whether we have an affiliate relationship with that brand.
We do not accept sponsored content, paid product placements disguised as editorial, or commission-based "best probiotic" lists. The distinction between our evidence ratings and our affiliate recommendations is maintained explicitly: a product with Limited Evidence may still be something we link to with full affiliate disclosure, but we will not inflate its evidence tier to generate clicks.
This site does not yet have a named registered dietitian or gastroenterologist as an editorial reviewer (a long-term goal to enhance E-E-A-T). Our content is research-synthesised by the Digital Signet editorial team with continuous reference to primary sources. We are a consumer-facing reference site, not a clinical decision support tool. For individual medical decisions, consult a qualified clinician.
Last Updated and Freshness Policy
All pages display a "Last updated" date. We review and update evidence-sensitive claims when new systematic reviews or Cochrane updates are published, when ISAPP publishes new consensus documents, or when major new RCTs change the evidence landscape. The 2024 IBS network meta-analysis (Ford et al., Am J Gastroenterol 2024) is the most recent major update incorporated into our strain recommendations.
If you see a claim that appears out of date or inconsistent with recent published evidence, contact us at contact@digitalsignet.com.