HL7 v2 MRG-1 (Prior Patient Identifier List): Format, Values, and Testing Notes
Get MRG-1 backwards and a merge feed doesn't fail loudly — it quietly retires the wrong medical record number while the audit log reports success.
MRG-1 is a repeating field holding the medical record number (and any other identifiers sharing that namespace) that used to belong to the losing patient record in a merge. It's the field a receiving system reads to figure out which MRN just got absorbed into the patient named in PID-3 of the same message. Because it can repeat, one MRG-1 field can carry more than one prior identifier when a duplicate record itself had multiple MRNs across facilities.
What this field contains
Example
This synthetic MRG-1 value, 847213609^^^HOSP^MR, is the retiring MRN being folded into the surviving patient identified elsewhere in the same message.
Where this trips people up
Teams frequently assume MRG-1 always contains exactly one identifier, then build parsers that only read the first repetition. When a source system sends a duplicate record that itself had two MRNs — one from a legacy registration system and one from the current EHR — the second repetition gets silently dropped, and that orphaned ID never gets reconciled against the surviving record. Test fixtures should include both single-value and multi-repetition MRG-1 fields.