CPCSSN does NOT extract all patient data. CPCSSN extracts mainly structured data (e.g. lab values, blood pressure) and not notes or PDFs contained within EMR. CPCSSN does NOT extract identifiable information—we take great pains to extract only de-identified data.
Data are extracted from specific fields (e.g. billing, health condition, reason for visit, lab, medications) and all patients are provided with a random CPCSSN ID.