How to Evaluate a Buried-Cure Claim

A buried-cure claim usually arrives with three parts: a treatment, a villain, and a promise. The treatment may be a plant, drug, diet, device, procedure, or old medical theory. The villain may be "Big Pharma," regulators, journals, universities, or a professional guild. The promise is usually stronger than the evidence: cured cancer, reversed disease, prevented infection, or exposed a secret.

The goal is not to become cynical. The goal is to slow the claim down until it can be checked. A good evaluation asks what would be true if the treatment worked, what evidence should exist, and whether the best available evidence matches the strength of the sales pitch.

The six-question test

1. What exact disease and outcome is claimed? "Boosts immunity" is not the same as preventing hospitalization. "Helps cancer" is not the same as shrinking a specific tumor type or improving survival.
2. What is the best human evidence? Case reports can suggest hypotheses. Randomized trials and systematic reviews carry more weight for treatment effects.
3. Was the negative evidence searched for? Look for trial registries, FDA records, Cochrane reviews, and professional summaries, not only testimonials.
4. Does the mechanism fit the dose and route? A compound can kill cells in a dish at concentrations that are irrelevant or unsafe in humans.
5. Who benefits from belief? Pharmaceutical incentives matter, but supplement sellers, clinics, influencers, and documentary makers also have incentives.
6. Has the claim narrowed over time? Honest evidence often narrows a claim. Promotional evidence tends to expand it.

Tool

ClinicalTrials.gov: check whether a claim has entered human testing

ClinicalTrials.gov is a registry and results database for clinical studies. It helps answer whether a therapy is being tested, what outcomes are being measured, whether the study is randomized, and whether results have been reported. A registered trial does not prove efficacy, but it is much stronger than a claim that "no one is allowed to study this."

Source: ClinicalTrials.gov study basics

Tool

PubMed and PMC: separate literature from marketing

PubMed indexes biomedical citations, and PubMed Central hosts full-text articles. Search the treatment name with the exact condition, then add terms such as randomized, systematic review, meta-analysis, toxicity, and adverse events. Do not stop at a mechanistic paper if the public claim is a clinical cure.

Source: PubMed about page

Tool

Cochrane: look for systematic reviews and plain-language summaries

Cochrane reviews are designed to find, appraise, and synthesize multiple studies. They are especially useful when individual papers point in different directions. A therapy with many small positive trials but a weak systematic review should be treated as unresolved, not proven.

Source: Cochrane evidence overview

Tool

FDA expanded access: compassionate use is not proof

Expanded access, sometimes called compassionate use, can allow a seriously ill patient to receive an investigational product outside a clinical trial when alternatives are lacking. That access pathway does not mean FDA has found the product safe and effective for the condition. It is a risk-managed exception, not a verdict.

Source: FDA expanded access overview

Evidence strength table

Evidence type	What it can show	What it cannot show alone
Historical use	A remedy was known, valued, or repeatedly tried.	That it works for a modern diagnostic category.
Mechanism or cell study	A plausible biological pathway worth testing.	Safe, effective treatment in humans.
Case report	A signal, adverse event, or unusual outcome.	Typical benefit, causality, or survival advantage.
Randomized controlled trial	Whether a defined intervention beat a comparator on defined outcomes.	Universal effectiveness across diseases, doses, and populations.
Systematic review	The shape of the evidence across multiple studies.	Certainty if all underlying trials are small, biased, or inconsistent.

Red flags

Be cautious when a claim promises one remedy for many unrelated diseases, treats every negative result as evidence of suppression, relies on testimonials while dismissing trials, sells a product or clinic package before evidence is clear, or tells patients that ordinary clinicians cannot be trusted under any circumstances.

Also be cautious when critics overstate their case. "No evidence" is different from "negative evidence." "Not enough evidence yet" is different from "impossible." The most useful position is precise.

Apply the test

Try the checklist on H. pylori, laetrile, and phage therapy. Then use the claim evaluator for a claim you are researching. The categories should become clear: one changed the causal model of ulcers, one failed the cancer-therapy claim, and one remains an active but not fully settled field.