ICD-10 Cm
Detects ICD-10 Cm patterns. This pattern is based on a Microsoft Purview built-in sensitive information type. Users already running Purview may prefer to enable the built-in SIT directly, or use this version as a starting point for customisation.
- Type
- regex
- Engine
- universal
- Confidence
- high
- Confidence justification
- High confidence: structurally constrained pattern with corroborative keyword support reduces false positive rates significantly. Added context gating and exclusion rules improve precision and reduce incidental matches.
- Detection quality
- Verified
- Jurisdictions
- global
- Frameworks
- ISO 27001, ISO 27701, SOC 2
- Data categories
- phi, healthcare
- Scope
- narrow
- Risk rating
- 8
- Platform compatibility
- Purview: Compatible, GCP DLP: Compatible, Macie: Compatible, Zscaler: Compatible, Palo Alto: Compatible, Netskope: Compatible
Pattern
\b[A-TV-Z]\d[A-Z0-9](\.?[A-Z0-9]{0,4})?\bCorroborative evidence keywords
MRN, medical record number, patient ID, NPI, DEA, medicare, medicaid, insurance ID, member ID, beneficiary, ICD-10, ICD-9, CPT, NDC, SNOMED, HCPCS, diagnosis code, procedure code, drug code, field (+28 more)
Proximity: 300 characters
Should match
A01— Cholera category code (3-char minimal)E11.65— Type 2 diabetes with hyperglycemiaS72.001A— Fracture of femur, initial encounter
Should not match
U01— Invalid first letter (U not in A-T, V-Z range)123— Starts with digit instead of letterA0— Too short (only 2 characters)template example placeholder record identifier— Template/sample context should be excluded even when anchor words are present
Known false positives
- Medical terminology in health education materials, research publications, clinical guidelines, or public health documents without patient-specific data. Mitigation: Require corroborative evidence keywords confirming patient context. Look for co-occurrence with patient identifiers such as medical record numbers or dates of birth.
- General wellness and fitness content using medical vocabulary without constituting protected health information. Mitigation: Layer with patient identifier patterns or healthcare-specific document structure detection to distinguish clinical records from general health content.