What FindMySite is
FindMySite is an independent research-planning interface for public ClinicalTrials.gov study-location records. It does not create or independently verify the source records. ClinicalTrials.gov explains that study sponsors and investigators are responsible for submitted information and that the U.S. government does not review or approve the safety and science of every listed study.
Read the official ClinicalTrials.gov disclaimer ↗.
How a search is produced
- Define the source cohort. The submitted phrase is sent to the ClinicalTrials.gov API v2 BasicSearch field. Country and city, when selected, are applied to study locations. Sponsor includes the lead sponsor and named collaborators; trial status is the study's overall status.
- Retrieve and reconcile. FindMySite walks every API page within its published guardrails, requires unique valid NCT identifiers, reconciles the fetched total to the source total, and checks that the source version remains stable. If those checks fail, no partial cohort is presented as complete.
- Derive facility candidates. The reconciled records are grouped and ranked in the user's browser. FindMySite does not maintain a ClinicalTrials.gov database or registry mirror.
A facility candidate is not a verified site identity
ClinicalTrials.gov does not supply a durable physical-site identifier. FindMySite groups a specific facility name, city, and country after Unicode normalization, case folding, and punctuation/spacing normalization. This is a deterministic convenience grouping, not legal-entity resolution, address verification, or proof that two similarly named records refer to the same operating site.
Generic labels and records without sufficient geography remain source-record-specific. Known contact-instruction text mistakenly published as a facility name is excluded from facility candidates.
How ranking works
Candidates are ordered by the number of distinct matching studies, then by studies with an active location status, then alphabetically by facility and city. “Active” means the facility location is Recruiting, Not yet recruiting, Active not recruiting, or Enrolling by invitation. It is kept separate from the study's overall status.
What the displayed fields mean
- Matching studies
- Distinct studies in the current reconciled search cohort linked to the candidate grouping.
- Published location records
- Source Location objects, which can be more numerous than studies.
- Started since (recent starts)
- Studies linked to the candidate whose registered start date is within the last three years of the retrieval month, counted over the whole reconciled cohort. Studies with no stated start date are counted in neither direction and are listed separately in the CSV. A start date is not evidence that the facility enrolled anyone.
- Terminated or withdrawn
- Studies linked to the candidate whose overall status is Terminated or Withdrawn, counted over the whole reconciled cohort. The registry does not say whether the facility caused a study to stop, so this is context, not a performance rating. Neither signal changes the ranking.
- Latest returned update
- The most recent ClinicalTrials.gov study-record update date among the evidence currently returned for that candidate. It is not a facility's own verification date or latest operational activity.
- Confidence (grouping)
- Shown on each facility candidate. It says how the facility was grouped, not its quality or capacity. Medium means registry entries share the exact same facility name, city and country; this is a name match, not a verified site identity. Low means the entry is listed on its own because the facility name or place is too generic to group with others.
- Registry map coordinate
- A source-supplied coordinate. When variants exist, FindMySite uses the most frequently reported one. Coordinates are not independently address-verified.
Full site portfolio
The full portfolio runs a separate facility search and only shows a study after a returned Location object exactly matches the selected normalized facility name, city, and country. It spans specialties and can therefore be broader than the disease cohort that produced the original candidate. Name changes, aliases, spelling differences, mergers, and incomplete registry locations can still split or omit a real-world institution.
Completeness controls and limits
- Maximum 50,000 studies per interactive source cohort.
- Maximum 125,000 location records per aggregation.
- Maximum 75,000 facility candidates per aggregation.
- Up to 500 ranked candidates are shown per page; headline totals continue to describe the reconciled cohort.
- API pages can be held in a five-minute Cloudflare edge cache. Each result displays the ClinicalTrials.gov source timestamp used for its reconciliation.
These are explicit resource guardrails, not silent sampling. A cohort that exceeds them at the global level is shown as complete country-level coverage, not as a partial facility ranking. Selecting a country begins a separate full reconciliation within that geography. If the selected geography still exceeds the guardrails, FindMySite asks for a narrower city, sponsor, status, or indication.
Suggested publications
Each study card can show papers suggested for that study. They are found by the study's NCT number only, read live from PubMed and from the references the sponsor lists on ClinicalTrials.gov, and nothing is stored. The rules match those of Physiolayer's MedAtlas (rules version 3):
- A reference the registry marks as a result or derived publication is linked.
- A paper that carries the NCT number as a PubMed registration number, or names it in its title or abstract, is linked.
- Reviews, meta-analyses, corrections, and papers naming more than three studies are not counted. A paper published more than two years before the study began is not counted. The panel reports how many such papers were set aside.
- Publication order alone never creates a link, and no language model is used.
- At most 100 PubMed results are read per request; when a study has more, the count is shown as a minimum (for example “12+”). Answers can be held in a six-hour edge cache.
These are suggestions from identifiers, not verified links. A paper that does not carry the NCT number will be missed, and a paper that mentions it without reporting on the study can be included. Read the paper before relying on it.
Required independent verification
Public registry history cannot establish current investigator availability, patient access, enrollment performance, startup time, contracting capacity, data quality, competing workload, regulatory readiness, or protocol fit. Confirm the current ClinicalTrials.gov record, contact the institution, and perform appropriate medical, operational, regulatory, privacy, and legal due diligence before making a site-selection decision.
Source and modifications
Source data: ClinicalTrials.gov ↗; suggested publications also use PubMed ↗. FindMySite modifies presentation by selecting fields, reconciling pages, normalizing facility labels and geography, grouping location records, calculating counts, ranking candidates, and selecting a map coordinate. These transformations are FindMySite's and are not official ClinicalTrials.gov classifications.