HL7 Table HL70189: Ethnic Group

HL70189 is the HL7-defined code table for ethnic group, aligned with the U.S. Office of Management and Budget (OMB) two-category ethnicity model (Hispanic or Latino vs. Not Hispanic or Latino) plus an Unknown placeholder. It sits behind PID-22 Ethnic Group on patient demographic messages. PID-22 is a repeating field, but ethnicity is effectively a single-cardinality concept under OMB rules; most senders populate one value per patient.

Unlike HL70005 Race, HL70189 uses single-letter codes (H, N, U) rather than the CDC numeric-hyphenated codes — though modern profiles often substitute the CDC equivalents (2135-2, 2186-5) directly into the same CWE field.

Purpose

HL70189 records whether a patient identifies as Hispanic or Latino — a separate question from race under OMB rules, which is why ethnicity has its own code table and its own PID field. The distinction matters for federal demographic reporting, Meaningful Use, USCDI, and health-equity analytics, all of which treat race and ethnicity as orthogonal dimensions.

Where it's used

• PID-22 Ethnic Group — primary home of HL70189; the field is repeating but typically populated with a single value.

Code list

Code system OID

• HL7-internal OID: 2.16.840.1.113883.18.66 — assigned to v2-0189 in HL7 Terminology.
• CDC Race & Ethnicity Code Set OID: 2.16.840.1.113883.6.238 — the upstream source for the CDC equivalents (2135-2, 2186-5) used in US Core.
• Canonical URI (HL7): http://terminology.hl7.org/CodeSystem/v2-0189
• Canonical URI (CDC): urn:oid:2.16.840.1.113883.6.238

HL7-defined vs user-defined

HL70189 is HL7-defined. The three letters above are the normative HL7 set. Sites that need granular ethnicity (specific national or cultural origins, e.g. Mexican, Cuban, Puerto Rican) do not extend HL70189 — they populate PID-22 with the full CDC Race & Ethnicity Code Set, citing the CDC OID in CWE.14 and letting the receiver roll up to H or N as needed. This is the same OMB-rollup pattern used for race.

Version differences

• v2.3.1 — HL70189 introduced as the user-defined Ethnic Group table.
• v2.5 — Re-anchored on the OMB two-category model; HL7-defined with H, N, U. PID-22 data type widened from IS to CE.
• v2.7 — PID-22 widened to CWE, enabling OID metadata in CWE.14 and supporting CDC numeric codes (2135-2, 2186-5) directly in CWE.1.
• v2.8 – v2.8.1 — Stable. No additions.

Common mistakes

• Putting ethnicity in PID-10 (Race) or race in PID-22 (Ethnic Group). OMB treats them as orthogonal; mixing them up breaks demographic reporting.
• Defaulting unknown ethnicity to N (Not Hispanic or Latino). U exists for unknown; defaulting to N introduces measurement bias in health-equity analytics.
• Treating H and 2135-2 as different concepts. They are the same concept in different code systems — the HL7 short letter and the CDC numeric code. Engines should normalize to one canonical representation.
• Sending the display text Hispanic in CWE.1 instead of the code. CWE.1 carries the code, CWE.2 carries the display.
• Extending HL70189 locally with letters like L or X. Granular ethnicity belongs in the CDC code set, not in custom HL70189 extensions.

Examples

A PID with ethnicity H (Hispanic or Latino):

PID|1||10456^^^MRN^MR||GARCIA^MARIA^A||19850412|F||2106-3^White^HL70005|123 MAIN ST^^BOSTON^MA^02118||||||||10456|||H^Hispanic or Latino^HL70189

Same patient with the CDC numeric code in CWE.1 and the HL70189 letter as alternate:

PID|1||10456^^^MRN^MR||GARCIA^MARIA||19850412|F||2106-3^White^HL70005|...||||||||10456|||2135-2^Hispanic or Latino^CDCREC^H^Hispanic or Latino^HL70189

Same H value translated to a FHIR US Core Ethnicity extension:

{
  "url": "http://hl7.org/fhir/us/core/StructureDefinition/us-core-ethnicity",
  "extension": [
    {
      "url": "ombCategory",
      "valueCoding": {
        "system": "urn:oid:2.16.840.1.113883.6.238",
        "code": "2135-2",
        "display": "Hispanic or Latino"
      }
    },
    {
      "url": "text",
      "valueString": "Hispanic or Latino"
    }
  ]
}

Mapping failure example — unknown vendor code:

PID|1||10456^^^MRN^MR||GARCIA^MARIA||19850412|F||...|||||||||10456|||LATINO^Latino^MERCYETH

LATINO is a vendor extension. A conformant engine should map it to H (and CDC 2135-2) via a curation rule, preserve the original MERCYETH code in a secondary Coding, and emit a curation alert.

FHIR mapping

There is no FHIR core ValueSet at HL70189's granularity. The standard target is the US Core Ethnicity extension (http://hl7.org/fhir/us/core/StructureDefinition/us-core-ethnicity), which carries:

• ombCategory — bound to the two OMB ethnicity codes (2135-2, 2186-5).
• detailed — bound to the full CDC Race & Ethnicity Code Set for granular Hispanic origin.
• text — a free-text representation.

The HL7 v2-to-FHIR IG provides field-level mapping for PID-22 to the US Core Ethnicity extension. U (Unknown) does not have a positive OMB equivalent; the conformant FHIR encoding omits ombCategory and uses a data-absent-reason extension instead.

Engine considerations

• Repeating field — PID-22 repeats with ~, but OMB ethnicity is single-cardinality; engines should treat additional repetitions as data-quality alerts rather than silently keeping the first or last.
• Letter-to-CDC normalization — H and 2135-2 are equivalent; engines must accept both shapes and normalize to one canonical internal representation.
• Absence vs Unknown — empty PID-22 means "not collected"; U means "asked and answered Unknown." Preserve the distinction in FHIR via data-absent-reason.
• Granular-to-rollup — when the source sends a granular CDC code (e.g. 2148-5 Mexican), compute the OMB rollup (2135-2 → HL70189 H) automatically.

How Vorro handles HL70189

Vorro accepts PID-22 in any of the common shapes — HL70189 letter, CDC numeric code, granular CDC code, or vendor extension. Vendor extensions route to the terminology curation queue, where Vorro either creates a remap rule (LATINO → H) or surfaces the original code as text while flagging for review. The CDC ethnicity hierarchy is loaded at startup, so granular codes automatically resolve to the OMB rollup.

On outbound, Vorro emits PID-22 as a single CWE value with both the HL70189 letter and the CDC numeric code populated across the primary and alternate triplets, allowing receivers to consume whichever system they expect. When ethnicity is genuinely unknown, Vorro emits U^Unknown^HL70189 and a corresponding FHIR data-absent-reason extension rather than collapsing to N.

Related pages

• PID segment — Patient Identification
• HL70005 code table — Race
• CWE data type — Coded With Exceptions

Sources

• HL7 Terminology — CodeSystem v2-0189 Ethnic Group
• HL7 v2 product brief
• CDC PHIN VADS — Race & Ethnicity Code Set