SMART Guidelines Starter Kit
2.0.0 - release International flag

This page is part of the SMART IG STARTER KIT (v2.0.0: Release) based on FHIR (HL7® FHIR® Standard) v5.0.0. This is the current published version. For a full list of available versions, see the Directory of published versions

L3 Authoring Overview

Authoring a L3 FHIR Implementation Guide involves taking the L2 input, using the recommended processes and tools, and creating, validating and publishing publishing L3 FHIR artifacts that correspond to the L2 specifications.

The L3 FHIR Implementation Guides are intended to be machine readable, but also will include narrative content that are human readable and shall speak to the L2 specifications covered in the FHIR IG being authored. More details about authoring narrative are included in this page.

Going from L2 to L3

The first step in L3 production is an input verification of the L2 content. L3 authors create the necessary artifacts to cover the entire L2 specification and demonstrate a working SMART Guideline. This includes several types of artifacts and possibly narrative content.

Input for authoring L3: L2 DAK components

For WHO published L2 DAKs, each DAK will be represented as a GitHub repository along with a published PDF DAK word document. The GitHub repository shall contain the structured spreadsheets, bpmn files and more, that were authored during the publication process of the DAK. Authoring presumes the L2 content is available and consistent. It is possible that the L2 content gets changed, notably due to input from L3 authoring. It is important to track the versions of L2 content, especially when adapting L2 or L3 to a local implementation, in order to track changes and assess impact of those changes.

Content Type L2 Representation L2 Location
Health Interventions Narrative Published DAK PDF
Generic Personas Narrative Published DAK PDF
User Scenarios Narrative Published DAK PDF
Processes BPMN files L2 DAK repository: input/business-processes/bpmn
Data Elements Spreadsheet L2 DAK repository: input/dictionary
Decision Tables Spreadsheet L2 DAK repository: input/decision-logic
DMN files L2 DAK repository: input/decision-logic/dmn
Scheduling Logic Spreadsheet L2 DAK repository: input/scheduling-logic
Indicators Spreadsheet L2 DAK repository: input/indicators
Functional Requirements Spreadsheet L2 DAK repository: input/system-requirements
Non-functional Requirements Spreadsheet L2 DAK repository: input/system-requirements

Authored L3 artifacts: Table of authored L3 artifacts as they correspond to L2 DAK components

The table and diagram below show the content types that are to be created as part of the L3 authoring process.

Content Type L3 Representation L3 Location Procedure
Health Interventions n/a n/a n/a
Generic Personas ActorDefinition input/actors Actors
User Scenarios ExampleScenario input/scenarios Scenarios
TestPlan input/testing Testing
TestScript input/testing Testing
Example instances input/testing
input/examples		 Examples
Processes FHIR PlanDefinition input/plandefinitions Plan Definitions
FHIR ActivityDefinition input/processes
input/activitydefinitions		 Activity definitions
Elements Logical Data Model input/models Logical Models
FHIR Questionnaire input/questionnaires Questionnaires
FHIR StructureMaps input/maps Structure Maps
FHIR ValueSet input/valuesets ValueSets
FHIR CodeSystem input/codesystems CodeSystems
FHIR Profile input/profiles Profiles
ConceptMap input/conceptmaps Concept Maps
CQL Concepts input/cql CQL
CQL Data Elements input/cql
Decision Tables FHIR PlanDefinition input/plandefinitions Decision Tables
FHIR Library input/libraries Concept Maps
CQL input/cql CQL
Indicators and Measures FHIR Measure input/measures Indicators
CQL IndicatorLogic input/cql CQL
Functional Requirements FHIR Requirements input/requirements Requirements
Non-functional Requirements FHIR Requirements input/requirements Requirements

Authoring L3 Artifacts: Where to begin?

While the exact authoring steps depend on the inputs, the key sequence could normally consist of the following steps: