Skip to main content
Version: v0.1.11

The three ways to read a message

HL7 v2 in 90 seconds​

HL7 v2 is a pipe-delimited messaging format used across US healthcare. A message is a sequence of line-oriented segments (3-letter names like MSH, PID, OBX). Each segment carries fields separated by |; fields decompose into components (^) and subcomponents (&). Repeating fields use ~.

A typical ADT message looks like this:

Message
 ├── MSH    (header: sender, receiver, type, timestamp)
 ├── EVN    (event details for ADT)
 ├── PID    (patient identification: name, record number, DOB)
 ├── PV1    (visit: class, location, attending)
 └── OBX×N  (observations: repeats for labs, vitals)

 Each segment   = pipe-delimited (|) fields
 Each field     = caret-delimited (^) components
 Each component = ampersand-delimited (&) subcomponents

That is enough HL7 to use this library. The HL7 v2.x standard reference has the rest.

There are three ways to reach into a parsed message, from the least HL7 knowledge to the most control. All three read the same immutable tree, so you can mix them in one function.

Named helpers​

Typed views over the common data. They need no HL7 knowledge, and they return undefined (never throw) when the segment they read is absent, so optional chaining reads naturally.

import { parseHL7 } from "@cosyte/hl7";

const msg = parseHL7(
  "MSH|^~\\&|LAB|MAIN|EHR|REF|20250102153045||ORU^R01|EX00080|P|2.5\r" +
    "PID|1||MRN12345^^^HOSP^MR||Smith^Jane^Q||19800115|F",
);

msg.patient?.mrn; // => "MRN12345"
msg.patient?.fullName; // => "Jane Q Smith"
msg.patient?.dateOfBirth?.precision; // => "day"
msg.meta.type; // => "ORU^R01"
msg.visit; // => undefined

Use them for most reads. The how-to guides, starting with Read patient and visit, cover each helper.

Dot-paths​

String paths in HL7's own SEG.field.component.subcomponent form. Field, component and subcomponent numbers count from 1, as the standard does. A bracketed index counts from 0: OBX[1] is the second OBX, and PID.3[1] the second repetition of PID-3. Without brackets you read the first.

import { parseHL7 } from "@cosyte/hl7";

const msg = parseHL7(
  "MSH|^~\\&|LAB|MAIN|EHR|REF|20250102153045||ORU^R01|EX00080|P|2.5\r" +
    "PID|1||MRN12345^^^HOSP^MR||Smith^Jane^Q\r" +
    "OBX|1|NM|GLU^Glucose^LN||120|mg/dL\r" +
    "OBX|2|NM|NA^Sodium^LN||140|mmol/L",
);

msg.get("PID.5.1"); // => "Smith"
msg.get("PID.5.2"); // => "Jane"
msg.get("OBX[1].5"); // => "140"
msg.getAll("OBX").length; // => 2

Reach for a dot-path when you want a field no helper surfaces, or for a quick script. A dot-path returns the decoded text of the first subcomponent it names; the helpers return typed values (the glucose result above is the number 120 from msg.observations()).

The segment tree​

Walk segments, fields, repetitions, components and subcomponents directly: for extractors the other two do not cover, and for edits.

import { parseHL7 } from "@cosyte/hl7";

const msg = parseHL7(
  "MSH|^~\\&|LAB|MAIN|EHR|REF|20250102153045||ORU^R01|EX00080|P|2.5\r" +
    "OBX|1|NM|GLU^Glucose^LN||120|mg/dL\r" +
    "OBX|2|NM|NA^Sodium^LN||140|mmol/L",
);

const rows = msg.segments("OBX").map((obx) => {
  const id = obx.field(3);
  const code = id.value; // OBX-3.1: value reads the first component
  const name = id.repetitions[0]?.components[1]?.subcomponents[0] ?? ""; // OBX-3.2
  return `${code} ${name} = ${obx.field(5).value}`;
});

rows; // => ["GLU Glucose = 120", "NA Sodium = 140"]
msg.segments("OBX")[0]?.field(3).asCwe().text; // => "Glucose"

seg.field(n) takes the HL7 field number, as a dot-path does. A Field's value is the first subcomponent of the first component of the first repetition; below it, repetitions, components and subcomponents are plain arrays counted from 0, so OBX-3.2 is components[1]. The typed coercions name the components for you: asCwe().text reads the same OBX-3.2, and asXpn(), asCx(), asXad() and the rest do the same for their types.

Every level is immutable. Changes go through explicit methods: see Modify a message.

Put it to use​