Blog 

Start with what the whole page promises

Before opening a footnote, write one sentence: “This page leads me to expect ___ from ___.” Name the outcome and the machine or treatment. Add any promise about speed, lasting results, home use or safety. This gives you something concrete to check rather than a vague impression that the page looks scientific.

US Federal Trade Commission guidance considers express and implied claims and the advertisement’s overall impression. Images and scientific presentation count alongside wording. A claim about the strength of evidence must itself be supported; a disclaimer cannot contradict the main claim. These are US advertising principles, not a ruling on a particular website or another country’s requirements. [1]

For your own audit, distinguish a claim that a treatment was studied from a claim that its advertised benefit was established. Then ask whether the sentence you wrote survives a close reading of the source.

On smaller screens, scroll the table sideways to read all three columns. This is a reader’s audit worksheet, not a legal compliance certificate. A missing answer means the claim has not been verified through that page.

A genuine citation can still be the wrong evidence

Open the original source and look for the sentence or result that supports the website’s promise. Do not stop at a familiar journal name. The three examples below show different ways the connection can break.

A comparative abstract does not identify every marketed machine

A trial indexed as PMID 26299702 included 60 participants in ultrasound-cavitation-plus-diet, cryolipolysis-plus-diet and diet-only groups. Its abstract reports waist and skinfold advantages for the treatment groups over diet alone, without a between-group weight or body mass index advantage. Only the abstract was verified for this guide; full text was not verified. It does not identify the precise frequency or model needed to match an advertised 40K or 80K machine. Nor does the absence of a significant difference between the active groups establish equivalence. [2]

A fat-layer finding is not automatically a waist-size result

The 2025 MED SmartCool study used a specified 1 MHz non-focused ultrasound device. Seventeen participants completed the prospective, uncontrolled study, with follow-up through two months. Researchers reported changes in superficial abdominal fat thickness, while abdominal circumference did not change significantly. A website that cites this paper to promise a guaranteed waist reduction has changed the outcome; one using it as proof for a generic 40K/80K home machine has also changed the device and setting. [3]

A thermal HIFU record is not generic cavitation validation

FDA record K100874 identifies the LipoSonix System Model 1 for non-invasive waist circumference reduction. Its summary describes high-intensity focused ultrasound (HIFU) producing thermal coagulation. That is a specific system and intended use, not proof for every device labeled “ultrasonic cavitation.” The document does not validate an unrelated low-frequency home unit. [4]

For each citation, record the device or intervention, people, measured outcome, follow-up and what the website adds. If details needed for the match are absent, write “not established from this source”. That is more accurate than either endorsing the claim or declaring that no relevant research exists.

Audit the images and the footnotes together

Look at the image with its headline and caption. Record whether it presents an individual example, a research image or an expected outcome. Then follow its attribution. Ask which device was used, when the photos were taken, whether other treatments were involved and whether the photograph belongs to the cited study. If the source is unclear, leave the image’s evidentiary status unresolved.

A useful distinction is authentic image versus supported message. Even with permission and an identified subject, the photo still leaves questions about attribution and typical results. In your audit, keep those questions open until the supporting evidence addresses them.

FTC guidance warns that fine print cannot repair a contradictory claim; disclosures must be clear and conspicuous. It also treats genuine consumer anecdotes as insufficient scientific substantiation for health-product effects. [1]

1. Read the asterisk immediately. Check whether it changes the device, population, outcome or timing named in the main message.
2. Find the original item. A complete paper title, DOI, PMID or regulatory number is easier to trace than “clinical studies show.”
3. Compare the image with the source. A study’s fat-thickness result does not by itself explain a large change in a marketing photograph.
4. Re-read the page on a phone. Note whether the limits are readable where the promise appears, rather than buried after a long gallery.

No clinical before-and-after image is reproduced here. The task is to verify the message made by an actual page, not to create an illustrative treatment result.

Separate regulatory identity from proof of the advertised result

An FDA badge needs an exact record behind it. FDA states that establishment registration and device listing do not denote approval, clearance or authorization; it does not issue device-facility registration certificates. A certificate-shaped image is therefore not enough to establish a device’s status. [5]

For a clearance claim, compare the device name, manufacturer and intended use in the official record with the page. K100874 is a 510(k) clearance for the named LipoSonix system, not a premarket approval for the entire cavitation category. Its waist-related scope cannot be transferred to another machine or a home-use claim. [4]

Keep two separate entries in your worksheet: “What is this device’s regulatory status?” and “What supports this advertised result?” Do not use a US record as a shortcut for CE status or UK/Australian requirements. Those need the relevant jurisdiction’s documents.

A worked example: a claim bigger than its footnote

Write the apparent promise as: “This home console has been shown to produce lasting circumference loss.” Now compare it with the actual citation: a specified 1 MHz device, an uncontrolled cohort with 17 completers, short follow-up, superficial fat-thickness findings and no significant abdominal-circumference change. The paper does not establish the fictional headline. [3]

A faithful description of that paper would identify the system and outcome, explain the uncontrolled design and limited follow-up, and stop there. It would not present the paper’s findings as a validated result for the pictured home console.

Record the conclusion as “the cited evidence does not support this page’s promise.” That is a focused assessment of the citation, not proof that the machine could never work, a fraud accusation or a legal ruling.

Common mistakes that make a proof claim look stronger

• Counting papers instead of tracing claims. Ten links are not ten relevant demonstrations. Mark which citation supports which promise.
• Accepting “studied technology” as an answer about the pictured device. Ask where the bridge between them is documented.
• Turning a selected outcome into a broad benefit. Keep waist circumference, fat thickness, weight, skin tightening and cellulite appearance as separate claims.
• Letting a reassuring qualifier end the audit. Return to the complete sentence you wrote about what the page leads you to expect.

A practical evidence request is: “Please identify the research supporting this result for this device and use, including the measured outcome, follow-up and relevant limitations.” Save the response beside the original claim. If it supplies only a category-level citation, the device-level promise remains unverified.

Use a claim-to-source test

Finish with a short record: the page’s promise, its cited support, the match or mismatch, and what remains unanswered. A convincing audit shows exactly where the connection holds or breaks. Scientific styling alone cannot do that work.