Note

Epic equivalent: Clinical documentation / Radiology reports / History & Physical / Discharge summaries

The note table captures unstructured clinical text — progress notes, discharge summaries, radiology reports, pathology interpretations, and other narrative documentation. In Epic, this maps to clinical documentation across all modules.

Each row is a single note or report. Downstream NLP pipelines can extract structured entities into the note_nlp table.

Epic-to-OMOP Field Mapping

Field reference (click to expand)

OMOP Field	Epic Equivalent	What It Captures
note_id	Note record ID	Unique identifier
person_id	Patient ID / MRN	Links to the patient
note_date	Note entry date	When authored; note_datetime has time precision
note_type_concept_id	Note category	Type: discharge summary, pathology report, progress note, etc.
note_class_concept_id	Note class	Format: dictation, transcription, structured note
note_title	Note header	Original title (e.g., "Follow-Up", "Operative Report")
note_text	Full note body	The unstructured text content
encoding_concept_id	Character encoding	Usually UTF-8 (metadata)
language_concept_id	Language	Typically English
provider_id	Author	Who wrote or dictated the note
visit_occurrence_id	Linked encounter	Visit context
note_source_value	Local note type	Original source label (e.g., "IM Progress Note")

What to Watch For

Common pitfalls

Large table — filter first

The note table can be massive. Always filter by date, note type, or visit before scanning. Check presence/absence of a note type for your patient sample before extracting full text.

Unstructured by nature

Notes require NLP to analyze at scale. The note_nlp table captures structured extractions.

Note titles vary by site and provider

Don't rely on note_title for classification — use note_type_concept_id instead.

Research Patterns

Question	Tables Involved
Advance directive documentation in free-text notes	note_text keyword/NLP search + note_type_concept_id
Discharge summary analysis for palliative care	note_text + condition_occurrence (palliative) + NLP
ECOG performance status from oncology notes	note_text + NLP extraction + condition_occurrence (cancer)
Suicidal ideation in notes vs. coded diagnoses	note_text keyword search vs. condition_occurrence
Most common note types in the ER	note_type_concept_id + visit_occurrence (ER filter)