DRAFT SMART Guidelines L3 SOP, published by TBD. This is not an authorized publication; it is the continuous build for version 0.2.1). This version is based on the current content of https://github.com/DigitalSQR/smart-ig-starter-kit and changes regularly. See the Directory of published versions
Decision Tables
Outputs:
- PlanDefinitions as part of the IG
- ActivityDefinitions as part of the IG or common repository
- Codes for the concepts used
Activities:
- Create a PlanDefinition for each Decision Table
- a. The PlanDefinition shall conform to the SMARTDecisionTable profile
- b. PlanDefinition ID should be the same as the decision table
- c. Title:
- d. Description:
- The output of the decision table should be one of the approved logical data objects that are part of the “actions catalog”.
- a. If an L3 author needs a new
- b. c
- For each Action in the decision table output:
- a. there should be a code of “standard” actions in the common or local repository*
- b. Each of those actions shall use a standard profile. Those standard profiles are also part of the activity outcome repository and respective governance.
- CQL Writing
- a. Ensure that every input for the CQL there’s a concept in the common dictionary
- b. See CQL criteria…
- c. Other elements
L3 artifact type catalog - getting tasks/immunizationrecom - cpg profiles
Guidance column - communication request
Actiona colum - machine action
Re: TB dak what happens when the action is just make clinical judgement
Outputs are nothing but go to the next step in the workflow or just present a smaller form with few options (TB testing type at facility)
Decision tree auto generation
Output Criteria / Definition of Done:
- Every Decision Table SHALL have a Requirements document pointing at it (for testability and inheritability)
- Every parameter in the decision table should be in the data dictionary
Change tracking
Known issues and dependencies:
