What is HL7?

Picture a hospital. Admissions runs on one software system, the lab on another, radiology on a third, the pharmacy on a fourth. None of them was built by the same company. Yet when a patient is admitted, they all need to know. When a result comes back, it has to land in the right record.

HL7 is the shared language they use to talk to each other. One system sends a small text message, another receives it and understands it, because both follow the same writing rules.

Not software, not a programming language

This is the classic trap. HL7 is not a program you install, nor a language like Python. It is amessaging standard: an agreement on how to structure information so a machine can read it without ambiguity.

If you know ICD-10 or CPT codes, you already have the intuition. "S06.0X1A" means exactly "concussion with loss of consciousness of 30 minutes or less, initial encounter", everywhere, for everyone. HL7 does the same, but for a whole message rather than a single diagnosis: it gives a shared structure.

What it looks like

Here is a real message: a sleep study result travelling from the lab to the patient record. Click any colored segment or any field. It explains itself.

At first glance it looks cryptic. That is expected: HL7 was designed to be read by programs, not by humans. The whole point of this site is to make it readable for you.

Three versions, only one running everywhere

You will run into three names. Here is how to keep them straight.

• HL7 v2: born in the late 1980s, recognizable by its vertical bars|. It is the version installed in almost every hospital in the world, still today. It is the one we learn first here.
• HL7 v3: an ambitious XML rewrite. Much heavier, rarely used in practice. You can set it aside for now.
• FHIR: the modern version, built on web technologies (REST, JSON). It is the future, and we cover it too. But FHIR does not replace v2 overnight: the two coexist, and understanding v2 makes FHIR obvious.