The Evidence Gradient

Wellness is full of confident claims. We would rather be trusted than impressive. The Evidence Gradient is the framework we use to match the strength of every claim we make to the evidence behind it, and it is the system behind the colored tiers on every page of this Library.

Practices built on light, field, and current sit in an unusual place. Some of their biology is standard physiology. Some of their clinical uses have been tested in trials that disagree. Some rest on models that have not been tested where they are offered, and some offer experiences people value with no clinical claim at all. A single phrase such as “evidence-based” cannot describe all four at once. The Gradient keeps them apart, so that the credibility of one never gets borrowed for another.

The four tiers and the floor

Each tier has an entry rule, a kind of language it allows, and a line it does not cross.

Tier I Established mechanisms

What qualifies. A biochemical or electrophysiological process replicated across laboratories and found in standard physiology.

How it sounds. “A mitochondrial enzyme absorbs red and near-infrared light.” “A changing magnetic field induces small currents in tissue.”

The line it does not cross. A mechanism never proves an outcome. That an enzyme absorbs red light is established (de Freitas & Hamblin, 2016); that red light restores your energy is a different claim entirely. Describing how something could work is not the same as showing that it does (Howick et al., 2010). See cytochrome c oxidase.

Tier II Mixed evidence

What qualifies. Comparative clinical studies exist, such as randomized trials or meta-analyses, but their results diverge across outcomes, populations, comparisons, or devices.

How it sounds. Specific findings credited to specific studies, with the disagreement stated: “one review found better fracture healing; a later review of acute fractures did not.”

The line it does not cross. A Tier II claim is never stated as settled. Most trials study clinical populations, so their results are indirect for someone seeking rest rather than treatment, the kind of indirectness that evidence-grading systems such as GRADE count against the strength of a conclusion (Guyatt et al., 2008). For wellness use, Tier II is a ceiling. See PEMF and TENS.

Tier III Theoretical biophysics

What qualifies. A coherent model grounded in credible science that has not been tested in the setting or for the purpose where it is being applied.

How it sounds. “One model proposes…” “Researchers are investigating whether…”

The line it does not cross. A Tier III claim is never presented as a finding or priced as a treatment. Bioelectric signaling is serious, well-published science in developmental biology (Levin, 2014); the leap from that science to a wellness session is a model, not a result. See bioelectricity.

Tier IV Experiential wellness art

What qualifies. Claims about calm, awe, presence, and rest in an immersive setting.

How it sounds. “Many people describe feeling calmer.” “The space is designed for rest.”

The line it does not cross. No disease claims, and no claims about the body’s physiology unless that physiology has been measured in that setting. Experience is worth studying in its own right: a randomized trial found that immersive relaxation reduced perceived stress more than the same content on a screen, though its authors note the trial was short and lacked follow-up (Nowakowska et al., 2026). See Portal of Light.

Below the floor Claims we do not make

What sits here. Claims that contradict well-established science or have failed direct tests. They are not treated as theories under investigation. They are simply not made.

An example. An independent review concluded that the Rife machine’s pathogen-specific frequencies, and its ability to kill pathogens, “are all unproven and highly implausible” (ALSUntangled Group, 2014, Conclusions section). A resonant light device offered as a quiet sensory experience can carry an honest Tier IV claim; the same device described as destroying pathogens falls below the floor. The device does not change. The claim does. See PERL.

How we place a claim

Each claim is checked against these questions in order, and it lands at the first one that fits.

1. Does it contradict established science, or has it failed a direct test? It goes below the floor and is not made.
2. Is it about a mechanism that is replicated and found in standard physiology? Tier I, with wording limited to the mechanism itself.
3. Have comparative clinical studies tested it? Tier II, with the disagreement and the population studied stated plainly.
4. Is it a credible model that has not been tested where we apply it? Tier III, labeled as a model.
5. Is it about experience: calm, awe, presence, rest? Tier IV, in the language of experience.

Rules we follow

• Rate the claim, not the device. One device can carry claims at several tiers at once, and each is labeled on its own.
• Mechanism does not license outcome. A Tier I fact supports only Tier I language.
• Discount for indirectness. Clinical evidence applied to wellness use is indirect, and we say so.
• Stay on the wellness side of the line. A wellness label describes intended use. Under the FDA’s general wellness policy, being covered by the policy does not establish that a product has been shown to be safe or effective for its intended use (U.S. Food and Drug Administration, 2026).
• Show the work. Every placement carries its citations, so you can disagree with a specific placement rather than with the practice as a whole.
• Date the placement. Evidence moves, and placements are revisited when new reviews appear.

Reading an entry

Every Library entry opens with its claims, one line each, with a colored badge for its tier. A term with claims at several tiers is not contradicting itself; it is being specific. Citations in a claim point to the references at the foot of the entry, and the entry text explains the findings in plain language.

The tiers are a sequence of claim types, not a strict ranking of certainty. Tier I is about mechanisms, Tier II about clinical outcomes, Tier III about untested models, and Tier IV about experience. A well-studied experiential finding can deserve more confidence than a weak clinical one.

What the Gradient is, and is not

The Evidence Gradient is a way of labeling public claims, assembled from existing work on evidence grading, mechanistic reasoning, and regulatory language. It borrows GRADE’s practice of keeping the quality of evidence separate from the strength of a recommendation, and its candor that any set of categories imposed on a continuum is somewhat arbitrary (Guyatt et al., 2008). It has not been tested as a measurement tool: no study has yet checked whether different people would place the same claims in the same tiers, and the placements in this Library are our own judgments, made by a practice rating its own offerings. Publishing the reasoning and the sources is how we hold ourselves to it. If you think a placement is wrong, we would like to hear why.

The last tier is not a weakness. Art that helps someone slow down, breathe, and feel present has value on its own terms.

Read the full framework in the Journal: The Evidence Gradient

References