Evidence First Decision Framework
This page explains how Evidence First Wellness evaluates research quality, interprets uncertainty, and distinguishes stronger evidence from weaker evidence. Explore the Framework →
Evidence Standards

The Evidence Standards
Behind EFW.

This page outlines how evidence is selected, interpreted, and applied across Evidence First Wellness. Not all research is equal in quality or relevance — and understanding that distinction is central to evidence-informed decision-making.

Why These Standards Exist

The supplement industry often presents
evidence differently than it is.

These standards exist to establish a consistent, transparent lens for how evidence is interpreted across this platform — not to create doubt, but to support better-calibrated decision-making.

What the industry often does
Presents preliminary evidence as definitive
Conflates biological plausibility with demonstrated benefit
Cites isolated studies as scientific consensus
Uses marketing language that implies clinical certainty
What these standards support
Interpreting evidence at the level it actually warrants
Distinguishing plausibility from demonstrated outcomes
Acknowledging uncertainty rather than resolving it artificially
Supporting calmer, more consistent supplement decisions

The goal is not to create perfect certainty. It is to support more thoughtful, evidence-informed decision-making.

Types of Evidence

Not all research is
equally informative.

Different study designs provide different levels of confidence. This site considers multiple types of evidence, with the weight assigned to each depending on study design quality and relevance to the question being asked.

Evidence hierarchy
Evidence weight
01
02
03
04
05
Systematic reviews & meta-analyses Synthesized evidence

Aggregates findings across multiple independent studies. Quality depends on the studies included — a meta-analysis of weak trials does not produce strong conclusions.

Randomized controlled trials (RCTs) Highest control

Participants randomly assigned to treatment or control groups, reducing confounding. The strongest design for establishing causation, though results do not always generalize across populations.

Cohort & longitudinal studies Observed over time

Follow participants over time without randomization. Can identify associations and track longer-term patterns. Cannot establish causation; confounders are a persistent limitation.

Cross-sectional & observational studies Association only

Observe patterns at a single point in time without controlling for confounders. Useful for generating hypotheses. Commonly the basis for supplement claims that overstate what the evidence supports.

Mechanistic & preclinical evidence Plausibility only

Cell and animal studies inform biological understanding. Do not confirm that effects translate to humans at relevant doses — a qualification frequently omitted in supplement marketing.

Tier placement reflects study design quality, not frequency of citation. Many studies at a lower tier do not accumulate into high-tier evidence.
Not all evidence types carry equal weight. The relevance of a given study depends on the question being asked — a mechanistic study may be highly informative for understanding a biological mechanism while being less informative for predicting a clinical outcome.
Strength vs. Relevance

A strong study can still
not apply to your situation.

A study can be methodologically strong and still not be relevant to a specific individual or context. Both dimensions matter when interpreting evidence.

Evidence strength

Refers to study design quality, sample size, and control for confounders. A large, well-designed RCT is stronger than a small observational study — but strength alone does not determine applicability.

Evidence relevance

Refers to whether the study population, dosage, duration, and outcomes apply to the context being evaluated. Research in elderly populations may not generalize to children. Clinical doses may not reflect supplement amounts.

Population specificity

Evidence from adults, athletes, or individuals with a diagnosed deficiency should not be assumed to apply to others. This site specifies the population studied when describing evidence.

Dosage and form

The dose and form used in research often differ substantially from commercial supplements. Evidence at a clinical dose does not automatically support the same outcome at a lower dose or in a different form.

Research Limitations

Limitations that are rarely
disclosed in marketing.

Most supplement research carries limitations that are not prominently disclosed in product marketing or popular summaries. This site acknowledges these limitations rather than presenting findings as more definitive than they are.

Small sample sizes

Many supplement studies involve dozens of participants. Findings from small samples are more susceptible to chance and less likely to replicate.

Short study duration

Effects observed over weeks may not reflect years of use. Short-term studies may miss both delayed benefits and delayed risks.

Surrogate endpoints

Many studies measure biomarkers rather than clinical outcomes. A change in a biomarker does not automatically translate to a meaningful health outcome.

Industry funding

Manufacturer-funded studies may be more likely to report positive findings — worth noting where independent replication is limited.

Publication bias

Studies with positive findings are more likely to be published, potentially presenting a more optimistic picture than the full evidence base supports.

Plausibility vs. Evidence

“Plausible” is not the same
as “demonstrated.”

Comparison framework
Evidence-supported Demonstrated in humans
vs
Theoretically plausible Mechanism, not outcome
Basis Observed outcomes in human clinical trials using relevant doses and populations
Basis Biological mechanisms from cell or animal studies, or extrapolation from related compounds
What it establishes That an effect was measured in specific people, under specific conditions
What it establishes That an effect is biologically possible — not that it occurs in humans at supplement doses
Common phrasing “In a 12-week RCT, participants showed…” / “A meta-analysis found…”
Common phrasing “Shown to support…” / “May help…” / “Research suggests…” — without specifying study type or population
Evidence-supported claims can still be narrow in scope. This site specifies who the evidence applies to, and at what dose.
Plausible claims are not dismissed — they reflect genuine scientific interest. But they are framed as preliminary, not proven.
Marketing vs. Evidence

Common patterns in
supplement claim language.

Marketing language in the supplement industry frequently extends beyond what the available evidence supports. Understanding common patterns helps identify when a claim is well-grounded and when it is overstated.

“Clinically studied”

Typically means the ingredient has been used in at least one study — it says nothing about study quality, the population studied, the dose used, or whether the finding has been replicated.

“Scientifically supported”

There is no formal threshold for what qualifies a supplement claim as scientifically supported. The phrase can be applied to evidence of any quality.

Structure / function claims

Under current US regulations, supplement manufacturers can make structure/function claims (e.g., “supports immune health”) without demonstrating that their product produces a measurable effect in users.

Proprietary blends

When ingredients are listed as part of a proprietary blend, individual amounts are not disclosed. This makes it impossible to assess whether ingredients are present at doses supported by evidence.

Absence of evidence

A lack of evidence does not confirm either effectiveness or safety — it indicates uncertainty. Where studies are limited or absent, this site acknowledges that uncertainty rather than interpreting it as reassurance or dismissal.

Where evidence is limited or mixed, that uncertainty is acknowledged explicitly rather than resolved artificially. This site focuses on helping readers understand evidence — not presenting supplement decisions as more straightforward than they are.
How a Claim Is Evaluated

A consistent sequence
behind every evaluation.

When assessing a supplement claim or study, this site works through a consistent sequence of questions — from the quality of available evidence to whether findings are practically meaningful. Uncertainty at any stage is acknowledged rather than carried forward as assumed confidence.

Evaluation sequence
Each step informs the next
Uncertainty noted at every stage, not resolved artificially
1

Identify the claim

What specific outcome is being claimed, for whom, and under what conditions? Many supplement claims are broad enough that they become difficult to evaluate as stated.

“What is actually being asserted here?”

2

Assess evidence quality

What study design supports this claim? How large was the sample? Was the study independently funded, and have findings been replicated?

“How strong is the evidence, and how was it generated?”

3

Evaluate consistency

Does the evidence replicate across independent studies? Mixed findings are common in supplement research and are disclosed rather than resolved in favor of either conclusion.

“Do multiple independent studies point in the same direction?”

4

Consider clinical relevance

Were the study population, dose, and duration applicable here? A statistically significant biomarker change is not the same as a meaningful clinical outcome.

“Does this evidence apply to the people and conditions being described?”

5

Gauge practical meaningfulness

Even where evidence is consistent, magnitude of effect matters. A small-but-real effect may not be meaningful relative to cost, dose, or available alternatives.

“If this is real, is the effect size meaningful in practice?”

Communicate what the evidence does — and does not — support

Claims are framed at the level of certainty the evidence warrants. Where uncertainty remains, it is stated rather than resolved with language that implies confidence.

These standards are applied consistently across articles, comparisons, and Digest issues. If you have questions about how a specific claim or piece of research has been interpreted, use the Contact page to submit a question — reader questions often inform future content.