Provider identifier records

Identifies healthcare provider identifier records - practitioner lists, credentialing files and directory extracts that carry labelled provider identifiers (Medicare provider number, prescriber number, HPI-I/HPI-O, Ahpra registration number, NPI, licence number) alongside practitioner and practice details. Format validation of each identifier is left to the dedicated identifier classifiers.

Type: regex
Engine: boost_regex
Confidence: medium
Confidence justification: Medium: provider identifiers are administrative and often public (registers, directories), so enforce tiers need a labelled identifier with practitioner/practice context and 85 needs two distinct identifiers plus credentialing evidence. Guidance and blank forms carry the labels without values and never match.
Detection quality: Topic false positive
Jurisdictions: global
Regulations: Healthcare Identifiers Act 2010 (Cth), Privacy Act 1988 (Cth), HIPAA, GDPR
Frameworks: ISO 27001, ISO 27701
Data categories: healthcare, phi
Scope: wide
Risk rating: 4
Platform compatibility: Purview: Compatible, GCP DLP: Unsupported, Macie: Unsupported, Zscaler: Compatible, Palo Alto: Unsupported, Netskope: Unsupported

Pattern

(?i)\b(?:(?:Medicare\s+)?provider\s*(?:no\.?|number|ID)|prescriber\s*(?:no\.?|number)|HPI-?[IO](?:\s+(?:no\.?|number))?|AHPRA\s*(?:registration\s*)?(?:no\.?|number|ID)?|registration\s*(?:no\.?|number)|NPI(?:\s+(?:no\.?|number))?|licen[cs]e\s*(?:no\.?|number))(?:[ \t]*\([^)\n]{1,20}\))?[ \t]*[:#][ \t]{0,3}(?!(?:n/?a|name|unknown|tbc|tba|withheld|not\s+recorded|the|a|an|any|all|each|see|refer)\b|_{2,}|\[|<)[A-Z0-9][A-Z0-9-]{3,17}(?:[ \t][0-9]{1,8}){0,3}

Corroborative evidence keywords

credentialing, scope of practice, conditions on registration, Medicare billing, Services Australia, NPPES

Proximity: 300 characters

Should match

Practitioner credentialing record Practitioner name: Dr Priya Ramanathan Specialty: General practice Medicare provider number: 2447181A AHPRA registration number: MED0001234567 HPI-I: 8003 6100 0012 3456 Credentialing complete; scope of practice approved for Medicare billing at this location. — Credentialing record with three labelled provider identifiers, practitioner fields and credentialing evidence (85)
Provider number: 2447181A Practitioner: Dr Jordan Staffson — Labelled provider number with a practitioner field (75)
Prescriber number: 1234567 — Labelled prescriber number alone (65 discovery)

Should not match

unrelated generic text — No relevant phrase context
placeholder value 12345 — Random text should not match phrase marker
generic narrative without identifier/document anchors — Should not match plain mention
template example placeholder record identifier — Template/sample context should be excluded even when anchor words are present
Policy guidance: this procedure explains how provider identifiers are recorded in the credentialing system and who may update them. It contains no identifier values. — Policy/guidance naming provider identifiers without values
Provider credentialing form Practitioner name: ____________________ Medicare provider number: ____________________ AHPRA registration number: ____________________ HPI-I: ____________________ Signature: ____________ Date: __/__/____ — Blank form with unpopulated labelled fields (placeholder guard keeps it below 75)
News: Services Australia will issue a provider number to new graduates within five days, and Ahpra registration checks remain free online. — News/training prose about the topic without a populated record (65 discovery at most)

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.