Skip to main content

Curate family history with pedigree detail

This recipe is part of Advanced Data Curation: capturing facts that do not exist in OMOP.

User goal
Structure family-history narrative, including relative, condition, age at onset, and genetic results.

When to use this
Use for hereditary-risk stratification, genetic-counseling referral worklists, or research cohorts that depend on family history.

Required inputs/data

  • Dataset built from a cohort or ingestion results (see the "Build a dataset" recipe reference in Scenario: S3 to cohort to dataset to Data Curation) for patients with social or family-history documentation
  • Note types such as H&P, new-patient/intake notes, genetics notes, and oncology notes
  • A custom registry defining the family-history fields

Custom registry fields (illustrative)

  • relative
  • condition
  • age_at_onset
  • vital_status
  • genetic_test_result

Patient Journey Intelligence features involved

  • Dataset Builder / Explorer
  • Data Curation Studio
  • Custom Registries
  • Curation Results and evidence review
  • Medical Terminology
  • Result export

Step-by-step workflow

  1. Build a dataset of patients with family-history documentation.
  2. Define a custom registry with the family-history fields, one row per relative-condition pair.
  3. Run the curation job.
  4. Review extracted relationships, conditions, and ages against evidence spans.
  5. Normalize condition terms where needed; correct and validate values.
  6. Export the structured family-history table for risk scoring or cohort building.

Expected output

  • Structured family-history records
  • Evidence-linked, reviewed values
  • Exportable table supporting hereditary-risk workflows

Edge cases / limitations

  • A single sentence can describe several relatives and conditions; model one record per relative-condition pair.
  • Negated history, such as "no family history of cancer," must not be captured as positive history.
  • Relationship terms can be ambiguous; preserve the source phrasing.
  • Self-reported family history is unverified and should be labeled accordingly for downstream use.

User value

  • Converts buried family-history prose into structured risk signals for screening, referral, and research.

Next step

See related recipes: Recipe 13: Curate reason for medication discontinuation or switch, Recipe 14: Curate implanted-device / UDI details, Recipe 15: Curate cancer recurrence events.