Home › Know Your Body

How to Spot a Bad Health Claim

By the Strength from the Well Editorial Team · Evidence last reviewed September 2026 · Not reviewed by a licensed clinician (why we say that)

The short answer

Bad health claims tend to share a structure rather than a subject: a dramatic promise, a story instead of a trial, a reason the evidence is missing, and something for sale at the end. The fastest check is to ask what specific outcome is being claimed, in whom, compared to what, and who benefits if you believe it. Weak claims almost always collapse on the second question.

Key points

The pattern matters more than the topic. The same template gets reused across supplements, devices, diets, and clinics.
A claim that covers many unrelated conditions at once is a marketing claim, not a medical one.
Testimonials feel like evidence and are not. They cannot show what would have happened anyway.
'Studies show' is not a citation. Ask which study, in whom, and what it measured.
Being skeptical of a seller is not the same as being skeptical of medicine. Cranks and marketers both profit from blurring that line.
Some unproven things later turn out to work. The right response is 'not established yet', not 'fraud' and not 'buy now'.

Why this is a pattern-recognition skill

You cannot evaluate every health claim on its merits. There are too many of them and they arrive faster than anyone can read. What you can do is learn the shapes that unreliable claims tend to take, because the people producing them are largely reusing a small number of templates.

This is not cynicism about medicine and it is not a claim that everything new is a scam. Plenty of legitimate treatments began as fringe ideas, and plenty of mainstream practices were later overturned. The goal is calibration: to spend your attention, money, and hope on the claims that deserve them.

The promise is too broad

Real treatments are narrow. A drug approved for a condition is approved for that condition, in a defined group, with a list of situations where it should not be used. That narrowness is a sign of seriousness, because it reflects what was actually tested.

So when something is claimed to help with fatigue, joint pain, brain fog, weight, sleep, immunity, and inflammation all at once, the breadth itself is the warning. Biology does not usually work that way. What is happening is that the claim has been widened to match the widest possible audience, which is a marketing decision rather than a scientific one.

A related version of this is the claim that a product addresses 'the root cause' of a broad set of problems. Root cause language sounds rigorous and usually is not, because it is applied to symptom clusters that have many different underlying causes in different people.

Stories where evidence should be

Testimonials are the most persuasive weak evidence there is, and understanding why they fail is worth a few minutes.

A person who took something and got better cannot tell you what would have happened if they had not. Many conditions fluctuate and improve on their own. People usually start something new when symptoms are at their worst, so the natural drift back toward average gets credited to the treatment. People who feel better are far more likely to write a review than people who felt nothing. And expectation genuinely changes how symptoms are experienced, which is a real effect and precisely why trials use comparison groups.

None of this means the person is lying. Most testimonials are sincere. Sincerity is not the issue. The issue is that a story has no comparison group, and without one there is no way to separate the treatment from everything else that was happening.

The evidence has an excuse attached

When a claim is strong and the evidence is thin, something has to fill the gap. Watch for the explanation offered for why the proof is missing.

The common form is suppression: it cannot be patented so nobody will fund a trial, or the industry is keeping it quiet. This is a closed loop. It explains the absence of evidence in a way that makes the absence feel like further confirmation, which means no possible observation could change the claim. That is the defining feature of an unfalsifiable argument, and it is the point where a discussion stops being about evidence.

It is worth being fair here, because there is a grain of truth underneath. Unpatentable and off-patent substances genuinely do attract less trial funding. That is a real gap in how research gets paid for. But a real funding gap produces small and inconclusive studies, not zero studies and a confident conclusion. If someone is certain, and the only reason offered for the missing evidence is that powerful people prevented it, the certainty is coming from somewhere other than data.

Citations that do not survive contact

'Clinically proven' and 'studies show' are phrases, not evidence. When a specific study is named, spend five minutes looking it up on PubMed. A surprising share of the time, one of these is true.

The study is in cells or mice. The study is in twelve people for three weeks. The study measured a different outcome than the one being claimed. The study tested a different form or a different substance entirely. The study found nothing, and the claim rests on a subgroup the authors flagged as exploratory. Or the study exists and says roughly what is claimed, but it is one study and later work did not replicate it.

This check is the single highest-yield thing you can do. It requires no expertise, only the willingness to look. Most claims are not built to survive anyone actually opening the citation.

Follow the money without becoming paranoid about it

Ask who benefits if you believe this. Not to dismiss them, but to know what you are reading.

If the most enthusiastic source is also the seller, the clinic, or an affiliate earning a commission, you are reading advertising. That does not automatically make it false. Companies do sell things that work. It does mean you should not treat their summary of the evidence as a neutral one, and you should look for a source with no stake before you decide.

Apply this symmetrically. A wellness influencer with a discount code has a financial interest. So does a clinic offering an expensive cash-pay infusion. So, in a different way, does a company funding a trial of its own drug. Noticing an interest is the start of the analysis, not the end of it.

The pressure tells on itself

Legitimate medical information is rarely urgent in the way a sales page is urgent. Nobody at the National Library of Medicine needs you to decide today.

Urgency, scarcity, and countdown timers exist to prevent the exact behavior that protects you, which is going away, reading, and coming back. If a health decision is being framed as a closing window, the window is usually the offer and not your health.

Watch, too, for pressure applied to your relationships. Advice to keep it from your doctor, or framing your doctor as too closed-minded to understand, is not protecting you from anything. It is removing the person most likely to catch an interaction with a medication you are already taking.

Limited supply, price rising, enrollment closing.
'Do not tell your doctor' or 'doctors will not understand this'.
A free consultation that exists to sell a package.
Payment plans for a treatment with no published evidence.
A refund guarantee used as a substitute for data.
Claims that a diagnosis you already have was wrong, from someone who has not examined you.

What a trustworthy claim looks like

It is worth knowing the positive pattern too, so you are not left only with suspicion.

A trustworthy source is specific about what the treatment does and does not do. It names the population studied. It reports the size of the effect rather than only its direction. It mentions side effects and who should avoid it. It distinguishes between what is established and what is still being tested. It updates when new evidence arrives, and says that it has.

Most importantly, a trustworthy source can tell you what would change its mind. That question, asked of any claim or any person making one, sorts things quickly.

The other failure mode: dismissing too fast

There is a mirror-image error, and it costs people as much. It is deciding that anything unfamiliar is nonsense, and that anything not yet offered by your own doctor must not be real.

Medicine changes. Treatments move from experimental to standard, and standard practices get abandoned when better evidence arrives. A physician being unfamiliar with something is not evidence that it does not work. It may only mean it is outside their specialty or newer than their training.

So the accurate middle position, and the one worth holding, has three categories rather than two: established, promising but unproven, and unsupported. Most things you will encounter belong in the middle category. Saying 'this is not established yet' is more honest than either enthusiasm or dismissal, and it leaves you able to change your mind when the evidence arrives.

Common questions

Does 'FDA approved' mean something is proven?

For drugs and devices it means the agency reviewed evidence of safety and effectiveness for specific uses. It is meaningful, though approval is for defined indications and does not mean the product is right for you. Dietary supplements are regulated differently and are not approved by the FDA before sale, so 'FDA registered facility' on a supplement label is not an approval of the product.

What about 'natural' products, are they safer?

Natural describes origin, not safety. Many potent drugs and many poisons are plant-derived. Supplements can also interact with prescription medications, which is why your clinician should know everything you take, including things bought without a prescription.

A doctor is making the claim. Does that change things?

It raises the baseline, but credentials do not transfer across topics, and a clinician selling a product has the same financial interest anyone else would. Look at the evidence offered and whether it matches the size of the claim, exactly as you would for anyone else.

How do I talk to a family member who believes a bad claim?

Arguing the conclusion rarely works. Asking questions does better: what specifically is it supposed to do, who did the study, what were they measuring. Curiosity keeps the conversation open, where contradiction usually closes it. Their underlying worry is generally real even when the claim is not.

Is there a quick check when I only have a minute?

Ask two questions: what exact outcome is claimed, in whom, and who profits if I believe it. Then search the product or substance name along with the word 'review' on a government or major medical center site rather than a search engine at large.