Why big site counts mislead: one site, many listings
A big site count sounds like coverage, but much of it is noise. The same research site is often listed many different ways across registries and databases, so one place looks like several. For a site plan, what matters is not how many sites a list holds, but how many are real, counted once and able to enroll your protocol.
- Registries and site databases list the same site under many names, addresses and departments. That is site sprawl.
- Some listings point to sites that closed, merged or moved. Those are ghost sites.
- Sprawl inflates site counts and can count the same patients twice.
- Ask how many sites are unique, current and proven on trials like yours.
- A short list of verified sites beats a long list of listings.
The problemOne site, many listings
Trial registries are filled in study by study, by many different people. A site’s name is typed in each time, so the same hospital can appear as the hospital, its research institute, a department, a clinic on another street or an abbreviation. Add mergers, rebrands, address moves and investigators who change institutions, and one real site can spread into a cluster of records. Site databases built on those registries inherit the problem. It has a name: site sprawl.
The costWhat noise does to a site plan
- Coverage looks bigger than it is. A list of thousands of sites can hold far fewer real places that could run your trial.
- Patients get counted twice. If one site appears twice, the patients it treats can be counted twice, and the enrollment plan starts from a number that was never there.
- Outreach hits the same site again and again. Sites notice, and it does not help the relationship.
- Ghost sites reach the shortlist. A listing for a site that closed, merged or stopped running trials still looks like a candidate. See ghost sites.
- The sponsor finds it first. In a bid defense, a duplicate or a dead site on your list costs more credibility than a shorter list ever would. See bid defense.
The testFive questions to ask about any site count
- How many of these are unique places, not listings?
- When was each site last confirmed as active and running trials?
- Which entries are the same site under another name?
- How many have run a trial like mine, and how did they enroll?
- How many treat patients who meet my protocol today? This one needs patient-level real-world data, not registry listings.
A provider who can answer all five has a site list. One who can only quote the total has a directory.
The shiftQuality beats quantity
Big numbers are easy to publish and hard to use. Everyone has access to the same registries, so a long list is not an edge. What wins a bid, and what keeps a trial on schedule, is a short list of proven sites: each one real, counted once, with patients who qualify under the protocol and an investigator with a record on trials like it. That is the idea behind verified site intelligence. Verified, not scraped.
Site counts, answered
Why are clinical trial site counts so high?
Because the same research site is often listed many times. Registries are filled in study by study, so one hospital can appear under different names, departments, addresses and abbreviations, and mergers and moves add more versions. Site databases built on those registries inherit the duplicates, so a big total often counts listings, not unique sites that can run a trial.
What is site sprawl?
Site sprawl is the way one research site multiplies across registries and databases, listed under different names, addresses and departments until it looks like several sites. It inflates site counts, can count the same patients twice and sends outreach to the same site more than once.
What is a ghost site?
A ghost site is a site that exists in the data but not in reality: a duplicate listing, a site that has closed, merged or moved, or one that no longer runs trials. Ghost sites inflate site lists, waste outreach and can reach a shortlist unnoticed.
Is a bigger site database better for site selection?
Not on its own. Size measures listings. For site selection, what matters is how many sites are unique, active and proven on trials like yours, and how many treat patients who meet your protocol. A short list of verified sites is more useful than a long list of unverified ones.
How do you check whether a site list is accurate?
Ask five questions: how many entries are unique places, when each site was last confirmed active, which entries are the same site under another name, how many have run a similar trial and how they enrolled, and how many treat patients who meet your protocol today. A list that answers all five is ready for a site plan.