How to read a ClinicalTrials.gov listing

ClinicalTrials.gov is run by the National Library of Medicine at the NIH, and it holds records for several hundred thousand studies from around the world. It is the first place most patients and families look when a doctor mentions a trial, or when they go searching on their own. It is also easy to misread, because every record follows the same official-looking layout whether it describes a large, carefully run trial or a clinic charging for an unproven procedure.

The site says so plainly in its disclaimer: "Listing a study does not mean it has been evaluated by the U.S. Federal Government." Sponsors and investigators write the content. NLM staff run limited quality checks, not a scientific review.

What follows walks through a record in roughly the order you meet it, then lists questions for the study team and the warning signs of listings that are really sales.

The layout of a record

A record page has a few tabs. Study Details is the plain summary most people read. Researcher View shows every registration field in a table, including ones the sponsor left blank, which is useful when you want to see what is missing. A results tab shows summary results if any have been posted. Record History lets you compare earlier versions, so you can see whether the main outcome or the size of the study changed along the way.

Section by section

The NCT number

Every record has an identifier made of "NCT" followed by eight digits. It is the single most useful thing on the page. Write it down. Use it to search PubMed and news coverage, quote it when you call the study team, and give it to your own doctor. The International Committee of Medical Journal Editors has required trial registration as a condition of publication since 2005, so a published trial should cite its NCT number.

Title and brief summary

There is usually a brief title in plain language and an official title in technical language. Read both. The brief summary is the sponsor's own description, so treat persuasive wording ("breakthrough", "regenerative", "natural healing") as a signal to read more carefully, not as information.

Recruitment status

Check the "last update posted" date next to the status. A record marked recruiting that has not been touched in years may not reflect reality.

Study type and phase

Interventional studies assign participants to a treatment. Observational studies watch and measure without assigning anything. Expanded access records describe ways to get an investigational product outside a trial.

Phases apply to drugs and biologics. Phase 1 studies are small and focus on safety and dosing. Phase 2 looks for a signal that the treatment works and collects more safety data. Phase 3 compares the treatment with standard care or placebo in larger groups, and Phase 4 follows a product after approval. "Not Applicable" usually means a device or behavioral study. Our explainer on how lab discoveries reach the clinic puts the phases in context.

Design

Four fields carry most of the weight. Allocation says whether people are randomized. Intervention model says whether groups run in parallel, cross over, or whether everyone gets the same thing (single group). Masking says who knows which treatment each person receives, from open label (everyone knows) to double or quadruple blind. Primary purpose says whether the aim is treatment, prevention, diagnosis or something else.

A single-group, open-label study with no comparison can tell you a fair amount about safety and very little about whether a treatment works better than doing nothing.

Arms and interventions

Each arm is a group: experimental, active comparator (an existing treatment), placebo comparator, sham comparator (a fake procedure) or no intervention. Look for what each group actually receives, how often and for how long, and work out your chance of landing in each arm.

Outcome measures

The primary outcome is the main question the study is built to answer, with a time frame. Secondary outcomes are extra questions. Good outcomes are specific and measurable: overall survival at two years, change in a named lab value at 12 weeks, number of serious adverse events. Vague outcomes such as "patient satisfaction" or "improvement in well-being", with no comparison group, are weak evidence however the study turns out.

Eligibility

Inclusion and exclusion criteria list who can and cannot take part: ages, diagnosis, prior treatments, lab values, other conditions. Meeting them on paper is only the start. The team will screen you, and many people who look eligible are not.

Sponsor, collaborators and oversight

The lead sponsor might be a drug company, a university, a hospital, a cooperative group or a government agency. Researcher View also shows oversight fields, including whether the study involves a US FDA-regulated drug or device and whether there is a data monitoring committee. A study run by the same business that sells the procedure, with one site and the owner as investigator, deserves extra questions.

Locations and contacts

Each site has its own status and often its own contact. The contact is usually a research coordinator, and that person is your best source for practical detail.

Dates

Look at study start, primary completion (when data for the main outcome is collected) and study completion. "Estimated" dates are plans; "Actual" dates are history.

Results

Under US law, many trials of FDA-regulated products must post summary results, generally within about a year of primary completion. Not every study is covered, so a blank results tab is not proof of wrongdoing. Posted results include participant flow, baseline characteristics, outcome data and adverse events. They are not peer reviewed, and they are often less polished than a journal paper, but they include the unflattering numbers too. Check the publications section for linked papers.

Questions to ask the study team

1. Which ethics board (IRB) approved this study?
2. Is it running under an FDA investigational new drug (IND) or device exemption (IDE)?
3. Will I pay anything? What does the sponsor cover, and what is billed to insurance as routine care?
4. Are travel, parking or lodging reimbursed?
5. What are the known risks, and what happens if I am harmed during the study?
6. How many visits, tests and procedures are involved, and over how long?
7. What is my chance of receiving placebo or the comparison treatment?
8. Can my own doctor stay involved, and will my results be shared with me?
9. What happens when the study ends? Will I have any access to the treatment?
10. How do I leave if I change my mind? (You can withdraw at any time.)

Spotting a pay-to-participate listing

Researchers have documented the problem for years. In 2017 Leigh Turner, writing in the journal Regenerative Medicine, described clinics registering "patient-funded" stem cell studies on ClinicalTrials.gov, and other studies where fees were charged but not mentioned in the record. The same year the New England Journal of Medicine reported three women with macular degeneration who suffered severe vision loss after eye injections of cells taken from their own fat at a Florida clinic. The procedure had been listed on ClinicalTrials.gov, and patients had seen the listing.

Warning signs, especially in combination:

• The record or the clinic's website mentions "patient-funded" or "self-funded" participation, or the contact number leads to a sales conversation about price.
• The sponsor is the clinic itself, there is a single site, and the investigator owns the business.
• One study covers many unrelated conditions, for example joint pain, lung disease, autism and multiple sclerosis.
• The intervention is the person's own fat or marrow cells, processed and injected the same day.
• The design is single group and open label, and the primary outcome is satisfaction or a loosely defined improvement.
• The study has run for years with no results and no publications.
• You are pressed to decide quickly or shown testimonials instead of data.

Legitimate trials rarely charge for the investigational treatment itself; in the US, charging for an investigational drug requires FDA authorization. The FDA's regenerative medicine consumer page covers unapproved stem cell products, and our guide to unproven stem cell clinics goes through the sales tactics. Clinicians may find our page on talking with patients about stem cell clinics useful.

FAQ

Does a ClinicalTrials.gov listing mean the FDA approved the study?

No. Registration means the sponsor entered information. Some listed studies are under FDA oversight and many are not, and a listing is not a seal of approval.

The study says "Recruiting." Will they take me?

Not necessarily. Recruiting means the study is enrolling, not that you qualify. Sites can also fill up or pause before the record changes.

Why is the status "Unknown"?

The sponsor has not confirmed the record in more than two years. Call the contact listed; the study may have ended or never started.

Can I join a trial in another country?

Some trials accept international participants, but travel, follow-up visits and insurance get complicated. Ask the team directly before planning anything.

Is it rude to ask about money?

No. Coordinators answer cost questions every day, and a legitimate team will explain what is covered in writing as part of informed consent.

What should I bring to my own doctor?

The NCT number and the eligibility criteria. With those two, your doctor can usually tell within a few minutes whether the study fits your situation and whether its design is likely to answer the question it asks.