HL7 v2 Reference Guides

The reference guides in this section cover the mechanics beneath every HL7 v2 interface: how values are typed, how messages are encoded, how they travel over the wire, how the standard evolved across versions, and how to diagnose interfaces when they misbehave. Message-level and segment-level detail live in their own sections; this is the layer underneath both.

Data types define what a field’s value looks like; encoding defines how those values are separated and escaped inside a message. Together they explain why PID-3 can carry an identifier, an assigning authority, and a type code in a single field.

MLLP (Minimal Lower Layer Protocol) wraps HL7 v2 messages for delivery over TCP. Nearly every production v2 interface runs on it, and most connectivity problems trace back to its framing or acknowledgment rules.

HL7 v2 is backward compatible by design, which is why a v2.3 sender and a v2.5.1 receiver can usually interoperate. Knowing what each version added tells you which fields you can rely on.

Organized by symptom rather than by cause, so you can start from what you observe (a growing queue, a silent socket, duplicate results) and work toward the fix.

Guide	Covers	Start here when
Data Types	CX, XPN, XAD, CE/CWE, TS, HD, PL, XCN	A field holds more components than you expected
Encoding	Delimiters, MSH-1/MSH-2, escape sequences	Carets or ampersands are showing up as literal text
MLLP	Framing bytes, ACKs, ports, TLS	Sockets connect but messages never arrive
Versions	v2.1 through v2.9, MSH-12	Sender and receiver disagree on a field’s meaning
Troubleshooting	Symptom-driven diagnosis	An interface worked yesterday and does not today