Case Study 2 — Twelve Fields, No Peptide
[constructed teaching example] The evaluation below is written as one reader's dossier entry. The compound-free claim is real and widely marketed; the entry is constructed to demonstrate method, and its ratings are this worked example's, dated, and yours to disagree with.
§40.8 made the argument that the twelve fields are not about peptides, and demonstrated it on a diet. This case study runs the harder version: a consumer medical device, where there is a manufacturer, a regulator, a marketing budget, a physical product, and — crucially — a regulatory label that says something true and is routinely read as saying something else.
The claim under evaluation: a consumer red-light therapy device — an LED panel or face mask emitting red and near-infrared light — improves skin appearance, relieves musculoskeletal pain, and aids recovery.
No molecule. No receptor. No vial, no sequence, no half-life, no injection. Watch the fields work anyway, and watch which failure modes turn up in exactly the places you have learned to expect them.
Field 1 — Identity
What is this, exactly?
"Red light therapy" is a name, not a specification — and this is Field 1's failure mode arriving on schedule, in a product category with no molecules in it at all.
A device is specified by parameters, and the ones that matter are:
- Wavelength(s), in nanometers. Red and near-infrared are different bands with different tissue penetration, and devices combine them differently.
- Irradiance — power delivered per unit area at a stated distance. This varies by more than an order of magnitude across consumer products.
- Treatment area and distance, since irradiance falls off sharply with distance.
- Whether the manufacturer publishes any of the above, and whether anyone independent has measured it.
Two devices sold under the same category name, with the same marketing language and the same red glow, can differ in every one of these. The category name is doing the same work "peptide" does in Chapter 1: borrowing credibility across a group whose members have little in common.
Same name, same thing? No. And, as with a research-labeled vial, that is a fact about the state of your knowledge rather than about the device.
Field 2 — Origin
Photobiomodulation has a long research history in dermatology, dentistry, and physical medicine, with a substantial laboratory literature and clinical use of professional-grade equipment.
Then note what that origin does and does not do. It establishes that this is a real research field rather than an invention of the wellness market — which is more than can be said for some claims. It establishes nothing about whether a particular consumer device produces a particular effect.
Record it; do not lean on it. The marketing leans on it constantly, in a form you will recognize: "backed by over 5,000 published studies." That is Field 5's failure mode — counting papers instead of reading designs — wearing Field 2's clothing.
Field 3 — Mechanism
In one plain sentence: red and near-infrared light penetrates a short distance into tissue, where it is proposed to be absorbed by mitochondrial components and to alter cellular energy metabolism and signaling.
Target: cytochrome c oxidase is the most commonly proposed chromophore.
Direct or indirect: indirect, and through several steps — light to absorption, absorption to cellular change, cellular change to a tissue-level effect, tissue-level effect to something a person notices. Every additional step is a place the chain can fail while every earlier step remains true.
Confidence in the mechanism itself: proposed. Parts are well supported in laboratory systems. Whether it operates at the doses a consumer device delivers through intact skin at a typical distance is a separate question, and one the mechanism cannot answer about itself.
And the standing rule: mechanism never upgrades a rating. Chapter 22's substance P antagonists bound their target exactly as designed and did not treat pain. A mechanism operating exactly as described in a dish is compatible with no effect at all in a person.
Field 4 — Pharmacology → dose and delivery
This field translates almost without modification, and it is where the evaluation turns.
For a drug you ask: what route, what exposure, and does the route it is actually used by match the route it was studied by? For a device you ask the same thing with different units. The exposure is defined by wavelength, irradiance, distance, area treated, and duration — and the relevant question is whether the exposure delivered by the product someone owns resembles the exposure delivered in the studies being cited to sell it.
Frequently it does not, for a reason worth stating plainly: much of the clinical literature used professional equipment at parameters that consumer devices do not publish and may not reach. A study conducted with a specified output at a specified distance is evidence about that exposure. A device that does not disclose its output is a device whose relationship to that evidence is unknown.
This is precisely Chapter 4's problem in a new costume. A peptide with oral human data being injected, or an injectable peptide sold as a cream, is a compound whose evidence base does not apply to how it is being used. The gap is invisible unless both are written down — which is why Field 4 has a line for each.
Route used matches route studied: unknown for most consumer products, and unknowable when parameters are undisclosed.
Field 5 — Evidence
The core field. Do not count studies; describe the best ones by design.
The design questions that decide everything here:
Was there a sham control? Sham devices exist — a unit identical in appearance and heat output but not delivering the active wavelengths. Trials with one are far more informative than trials without.
Was anyone blinded? This is genuinely hard. Visible red light is difficult to conceal from a participant, which pushes toward objective and investigator-blinded endpoints and away from self-report.
What was the endpoint, and who judged it? Investigator-rated appearance scales, standardized photography with blinded assessors, and objective measurements are different instruments from participant satisfaction questionnaires — and the gap between them widens exactly when blinding is imperfect.
Who ran it? A large share of the appearance literature is industry-associated. That does not make a study wrong. It is a reason to weight independent replication heavily.
Were the device parameters reported? A trial that does not state wavelength and irradiance cannot be applied to any specific product, including the one being sold on the strength of it.
Characterized honestly and without pretending to a precision the literature does not support: the skin-appearance literature contains randomized work, some of it sham-controlled, generally small, generally short, frequently industry-associated, with heterogeneous devices and endpoints. The musculoskeletal pain literature is larger, older, and strikingly heterogeneous in device parameters — which is itself the finding, because when the intervention varies that much, pooled analyses are combining things that are not the same intervention.
Conspicuously absent: large, independent, sham-controlled trials of specifically the consumer-grade devices being marketed, with pre-specified objective endpoints and durations beyond a few months.
Field 6 — Rating
One row per claim. Four rows, and the fourth is the one this whole case study exists for.
| Claim (population + endpoint) | Rating | Dated |
|---|---|---|
| Improves facial skin appearance; adults; investigator-rated or photographic appearance over weeks, using devices at reported parameters | ⚠️ | 2026 |
| Relieves musculoskeletal pain; adults; self-reported pain scores | ⚠️ | 2026 |
| Improves sleep, "recovery," or general wellness; healthy adults | ❌ (kind: evidence ABSENT) | 2026 |
| A specific undisclosed-parameter consumer device reproduces the effects reported in the clinical literature | ❌ (kind: evidence ABSENT) | 2026 |
Read the four rows together, because their relationship is the lesson.
The first two are genuinely ⚠️ — there is real human data and it does not settle the question. The third is ❌ with evidence absent, and it is the row that carries most of the advertising. And the fourth row is not about the therapy at all. It is about the product, and it is ❌ because the inference from "this modality has clinical evidence" to "this device delivers it" is a step nobody has taken with data.
That fourth row has no counterpart in most people's thinking, and it is the single most useful line in the entry. It is the same distinction as the one between a peptide and a vial: Chapter 32 established that identically labeled vials may differ materially, and Chapter 34 that a certificate describes a sample rather than the thing in your hand. A modality is not a product. Evidence attaches to specified exposures, and a device that does not specify its exposure has not inherited anything.
Note also that a single overall verdict on "red light therapy" — positive or dismissive — would destroy all four distinctions at once. That is rating rule 6, one intervention and many ratings, demonstrated on something with no chemistry in it.
Field 7 — Approved use
The field where this claim is most often misread, and the misreading is invited by the words.
Many consumer LED devices in the United States reach the market through 510(k) clearance, in which a manufacturer demonstrates that a device is substantially equivalent to a legally marketed predicate device. Some low-risk devices are exempt from even that. Higher-risk devices go through premarket approval, a different and far more demanding pathway.
So: "FDA cleared" is not "FDA approved," and neither is a finding that the device works for what it is being sold for. Clearance is a determination about equivalence to something already on the market. Where a cleared indication exists, it is often narrower than the marketing — and the cleared indication is the thing to read, not the letters "FDA" in the advertisement.
This is Chapter 38's argument arriving intact in a different regulatory regime: approval is a judgment about a specific claim, not a certificate about a product, and "not approved" has several distinct meanings that must not be collapsed. For most consumer wellness claims made about these devices, the honest entry is: no regulator has evaluated this claim, and none was required to.
Field 8 — Claimed use
Recorded in the seller's own terms, unedited: wrinkle reduction, collagen support, skin tone and firmness, pain relief, joint health, faster recovery, improved sleep, energy, mood, hair growth, "cellular health."
Overlap with Field 7: partial at best, and none for most of the list.
The gap between Fields 7 and 8 is, once again, the entire story — and it is visible only because they are separate fields. A single "what it's for" field would have merged a narrow cleared indication with a dozen unevaluated wellness claims into one impression, which is what the packaging does.
Field 9 — Risks
Four categories, kept separate.
At studied use: generally minor. Transient warmth, redness, occasional headache or eye strain reported in trials.
Specific to unstudied use: eye exposure is the real one. High-output devices should not be looked into, and eye protection is a manufacturer instruction to read rather than skip. Photosensitizing medications and conditions are a genuine interaction category and a clinician question, not a book question.
Specific to product quality: an undisclosed-output device is one whose exposure you cannot bound in either direction — you cannot confirm it is high enough to do anything or low enough to be trivial.
Unknown because nobody has looked: cumulative effects of daily use over years. There is no reporting pathway for a consumer wellness device comparable to post-marketing drug surveillance, so "no reported problems" describes the reporting system, not the safety profile — the same sentence you learned to write for an unapproved peptide.
And the honest summary: the risk profile here is genuinely mild compared with anything injected. That matters, and §40.9's point applies — it is an input to a decision, not the decision, and it does not raise the efficacy rating by a single tier.
Field 10 — Status
Five separate questions, and notice how differently they resolve from a peptide's.
Regulatory: consumer device; cleared or exempt rather than approved for efficacy. Sport: not a doping question. How it is actually sold: direct to consumer, across an enormous price range, with claims that frequently exceed anything in Field 7. Access and cost: freely available; the price spread between products with similar published specifications is large and not obviously explained by those specifications. Preparation quality → build quality and specification: this is where the peptide analogy lands hardest. There is no compounding risk, no sterility question, no counterfeit molecule. There is a different and structurally identical question: does the device deliver what its marketing implies, and can anyone independently verify it? Where parameters are undisclosed, the answer is no — which is Chapter 34's point about certificates, transplanted whole.
Keep this column entirely separate from Field 6. A device can be cheap, safe, widely available, and thinly evidenced. A device can be expensive, well-specified, and evidenced only for a narrow claim. Neither field adjusts the other.
Field 11 — Verdict
In one sentence, for my situation: for facial skin appearance there is real but preliminary human evidence supporting the modality at specified exposures, and no evidence that an undisclosed-parameter consumer device delivers those exposures; the wellness claims are unevaluated; the risk is low, which makes this a question about money and expectations rather than about safety.
Confidence: moderate for the modality, low for any specific product.
As a question a clinician could answer: "I'm on a medication — is photosensitivity something I should think about before using a red-light device regularly?" And for a dermatologist: "For what I'm trying to change about my skin, is this in the same category as the things you'd normally recommend, or a long way behind them?"
Both questions have answers that are in the room, which is §40.7's test.
Field 12 — What would change my mind
To move the appearance claim UP: an independent, sham-controlled, adequately powered randomized trial in a defined adult population, with blinded assessment of standardized photography as the pre-specified primary endpoint, at least twelve weeks, using a device whose wavelength and irradiance are published — and replication by a second independent group.
To move it DOWN: the same trial, reported null, which would convert this from evidence-preliminary to evidence-present-and-negative — a stronger state of knowledge than the one I am in now.
To move the fourth row — the product row — at all: independent measurement of consumer devices against their marketing claims, published, covering more than one manufacturer. This is the cheapest study in the entry and would resolve the most.
The study that would settle the wellness claims: it does not exist and no one appears to be attempting it, largely because "recovery" has not been given an endpoint anyone agrees on. Saying that plainly is more useful than any rating, and it is a well-specified open question rather than a shrug: name the endpoint and the study becomes designable.
Underway or planned? Independent device-measurement work: not that I can find.
What just happened
Twelve fields. No amino acids, no receptors, no half-life, no vial. And the same four errors turned up in the same four places:
- A category name mistaken for a specification (Field 1) — "red light therapy" doing exactly what "peptide" does.
- An origin story doing evidentiary work (Field 2) — a real research history recruited to support a specific consumer product.
- A mechanism standing in for an outcome (Field 3) — well-supported laboratory steps, an unverified chain to anything a person notices.
- A claimed use outrunning anything established (Fields 7 and 8) — a narrow clearance and a page of wellness claims, visible only because the fields are separate.
Plus one that the peptide chapters trained you for specifically and that most device buyers never consider: the difference between a modality and a product, which is the difference between a peptide and a vial, and which earned its own row in Field 6.
This is why §40.8 says the peptides were the curriculum. You did not learn facts about red light in this book. You learned where to stand when someone makes a confident claim about a complicated system, and the standing position does not change when the system does.
Questions
-
Field 6 contains four rows, and the fourth is about the product rather than the modality. Explain what that row is claiming, why it is rated ❌ evidence absent, and identify the two peptide chapters whose argument it reproduces.
-
"FDA cleared" and "FDA approved" appear in Field 7. State the difference precisely, then explain how it parallels Chapter 38's finding that "not approved" has several distinct meanings. Which meaning applies to the wellness claims in Field 8?
-
Field 4 concludes route used matches route studied: unknown, and unknowable when parameters are undisclosed. Construct the equivalent Field 4 line for a peptide product, and say what makes the two situations structurally identical despite having nothing physical in common.
-
Field 9 says "no reported problems" describes the reporting system rather than the safety profile. Explain the reasoning chain that has to hold for a "no reports" statement to carry information, and identify which links are missing for a consumer wellness device.
-
The entry rates two claims ⚠️ and two ❌ and refuses a single overall verdict. Write the single overall verdict a marketer would prefer, and then the one a dismissive skeptic would prefer, and state exactly what information each destroys. Which rating rule forbids both?
-
Choose a different non-peptide claim — a supplement, a procedure, a wearable, a dietary pattern — and identify, before doing any research at all, which three of the twelve fields you expect to be hardest to fill. Then fill those three. Was your prediction right, and what does the answer tell you about where you personally stop looking?