Build a Clinical Trial Site Dataset

Run condition or sponsor searches with the ClinicalTrials.gov Scraper, then explode each study's locations array. Carry NCT ID, overall status, condition, intervention, sponsor, and last-update date onto each site row.

Use a compound source key such as NCT ID plus facility, city, and country. Facility names are not guaranteed to be canonical, so aggressive merging can combine unrelated sites. Store the original label beside any normalized institution name.

Map coverage only after profiling missing coordinates and location status. A registered location does not necessarily mean it is currently enrolling. For operational use, verify the official record and listed contacts before outreach.