Open Nursing Core FHIR Implementation Guide (ONC-IG) - Local Development build (v1.0.0) built by the FHIR (HL7® FHIR® Standard) Build Tools. See the Directory of published versions

Structured Data Capture

Structured Data Capture

This IG ships SDC (Structured Data Capture) Questionnaires for its core nursing assessments, so ONC content can be rendered and captured directly by SDC-conformant form renderers — in particular the Open Health Stack data-capture libraries (android-fhir and kotlin-fhir-data-capture) used in offline-first deployments by Ministries of Health, NGOs and community services.

Available questionnaires

Questionnaire	Captures	Extracts toward
NEWS2	All seven physiological parameters + total score	NEWS2 Score and the vital-sign profiles
Monk Skin Tone	Skin tone on the 10-point Monk Scale (A–J)	Monk Skin Tone Observation
Braden Scale	Six subscales + total + mandatory Monk skin tone	Braden Scale Assessment
Waterlow Score	Total score + risk factors + mandatory Monk skin tone	Waterlow Score
MUST	Three step scores + total	MUST Score
Person-Centred Care	"What Matters to Me" + reasonable adjustments (repeating)	What Matters to Me, Reasonable Adjustment

How extraction works

The questionnaires use observation-based extraction (sdc-questionnaire-observationExtract = true at the questionnaire root):

• Every item that carries an item.code extracts to an Observation with that code. Display/guidance items and pure logic flags (e.g. the SpO2 Scale 2 toggle) carry no code and do not extract.
• Integer and decimal items carry the core questionnaire-unit extension, so extracted values become valueQuantity with the UCUM unit the ONC profiles require (e.g. {score} for assessment totals, /min, %, Cel, mm[Hg] for vital signs).
• Choice items (ACVPU, oxygen status, Monk skin tone) extract to valueCodeableConcept, with options drawn from the same value sets the profiles bind.

Implementer responsibilities

Observation-based extraction produces one Observation per coded item. Three things remain the capturing application's responsibility:

1. Category and performer. Set Observation.category to observation-category#survey and populate performer, as required by the ONC Nursing Assessment base profile.
2. Composite assembly. Braden and MUST define their subscales as Observation.component on a single resource. The questionnaires capture every component code, so assembling the composite Observation (or transforming via StructureMap) is mechanical.
3. The equity gate. Braden and Waterlow require a hasMember reference to a skin-tone Observation. Both questionnaires embed a mandatory Monk Skin Tone item, so the data is always captured in the same form response — the app links the extracted skin-tone Observation via hasMember[skinTone].

Equity by design, at the capture layer

The IG's fairness gate — no pressure-area assessment without a skin-tone record — is enforced here in the form itself: the Braden and Waterlow questionnaires will not complete without a Monk Skin Tone Scale rating. This means the equity constraint holds even in offline settings before any server-side validation runs.

Status

These questionnaires are published at draft status: they build and validate as part of this IG, and their extraction design has been verified against the target profiles, but they have not yet been demonstrated end-to-end in the Android FHIR SDK demo app. That verification step is tracked in issue #125 — implementers who render one of these forms are warmly invited to report results there.

---

Forms support clinical judgement; they do not replace it. Score interpretation and escalation decisions remain the responsibility of a registered nurse or other competent clinician.