Collaborative workflow
Snap2SNOMED is designed for collaborative map development. When creating a map, the user decides between two overall workflows, namely:
Single author mode
Dual author mode
Each of these overall approaches supports a range of options for collaboratively developing a map from a given source code set to SNOMED CT.
Single Author Mode
This overall approach supports multiple users working simultaneously on a map in a single author, single reviewer per source code manner. This means that work can be divided across a number of personnel to get work done.
A variety of workflows are currently supported, including:
Single user author + review workflow
A single user completes all author tasks for the map and then completes a review process on all mapped rows
Single user author (no review) workflow
A single user completes all author tasks for the map.
No review process is required
Single author + single review workflow
A single author completes all author tasks for the map.
A second user, the reviewer, will complete a review process on all mapped rows.
Multiple author + multiple review workflow
Two or more authors create mappings for each source code of the map.
Two or more reviewers completes the review processes on all mapped rows.
NOTE: A user can act as both an author and reviewer within the overall map (as assigned in a task), however, rules have been implemented to discourage the scenario of a user to reviewing their own authoring work. This can be overridden by an owner if required.
In the example flow shown, User 1 and User 4 can be the same person, ideally only reviewing the work of User 1+2.
Dual Author Mode
Dual author mode allows two authors to independently map the same source codes. Conflicts arising from differing interpretations are identified and resolved through a reconciliation process. This workflow brings the advantage of diverse perspectives, allowing for a comprehensive assessment of mappings. It ensures a cross-validation process where conflicting interpretations can be discussed, resolved, and reconciled, leading to accurate and reliable mappings between different coding systems or terminologies.
A variety of workflows are currently supported, including:
Two authors + single reconciler and single reviewer workflow
Two users user completes all author tasks for the map.
A reconciler completes the reconciliation of conflicting maps.
An assigned reviewer completes the review of all map rows.
Two authors + single reconciler and multiple reviewers workflow
Two users user completes all author tasks for the map.
A reconciler completes the reconciliation of conflicting maps.
One or more users are assigned as reviewers for a specific set of source codes. Each map row is assigned to a single reviewer.
Dual author mode with multiple authors + multiple reconcilers and multiple reviewers workflow
Two or more users complete the author tasks that have been assigned to them. The work is divided so that two users are mapping each source code.
One or more users are assigned as a reconciler. Each map row is assigned to one reconciler.
One or more users are assigned as reviewers for a specific set of source codes. Each map row is assigned to a single reviewer.
Two authors + reconciler and no review workflow
Two users complete all author tasks for the map.
A reconciler completes the reconciliation of conflicting maps.
No review process is required.
Review in Dual Author Mode
In Dual Author Mode, two independent mappers work separately on the same mapping task. The system keeps their mappings blinded until both have completed their work. Once both mappers submit their mappings, the system compares the results:
If both mappers select the same target, the row is marked as MAPPED and requires no further action.
If the mappers select different targets, the row automatically moves to RECONCILE, where a reconciler will review and determine the final mapping.
Key Considerations for Reviewing in Dual Author Mode
No Dedicated Reviewer Role: Unlike Single Author Mode, this mode does not require a separate reviewer because the reconciliation process ensures that discrepancies are addressed.
Modifying a Locked Mapping: Once a mapping is unblinded, it cannot be changed by either author. However, if adjustments are necessary, the status can be changed to RECONCILE using bulk change, or the Clear Targets option can be applied via bulk edit to reset the mapping and the blinding flag.
Handling Rejected Rows: If a reviewer (in an assigned review workflow) rejects a row, its status can be changed to RECONCILE using the bulk change feature. A reconciliation task can then be assigned to ensure the mapping is corrected.
Last updated