T3.4: Describes Worklist Filters and Search Options

Overlap and overlay are two distinct types of MPIID (Master Patient Index ID) conflicts that appear on the worklist. Understanding the difference is critical for the HS-EMPI Technical Specialist exam.

Overlap Definition: An overlap occurs when two records are linked together (the system believes they represent the same person), but they have different MPIIDs. This violates a fundamental rule of InterSystems EMPI: if two records represent the same individual, they must have the same MPIID.

How Overlaps Are Created: Overlaps typically occur when you assign a new MPIID to one record in a former duplicate situation, but a third record exists that was originally linked to both records in the duplicate. For example:

• Records A and B were previously linked as a duplicate (both had MPIID 12345)
• You unlink them and assign record A a new MPIID (67890)
• Record C was linked to the original MPIID (12345)
• Now record C is still linked to record B (MPIID 12345) but also linked to record A (MPIID 67890)
• This creates an overlap for record C

Overlay Definition: An overlay occurs when two records are NOT linked together (the system believes they represent different people), but they have the same MPIID. This also violates EMPI integrity rules: if two records have the same MPIID, they should be linked.

How Overlays Are Created: Overlays typically occur when you unlink a duplicate without assigning a new MPIID to one of the records. For example:

• Records D and E both have MPIID 11111 and are linked
• You unlink them using the "Different" button without using "A is Different" or "B is Different"
• Now both records still have MPIID 11111 but are no longer linked
• This creates an overlay

Resolution Procedures:

For Overlaps: Navigate to the Comparison Detail, Whole Record Viewer, or other Alternate Views, then:

• Click Same to link the records and synchronize MPIIDs (one record adopts the MPIID of the other)
• Click Different to unlink the records (each will maintain its current MPIID)

For Overlays: Navigate to the Alternate Views, then:

• Click Same to link the records (they already have the same MPIID, so this just creates the link)
• Click A is Different to unlink and assign record A a new MPIID
• Click B is Different to unlink and assign record B a new MPIID

---

The InterSystems EMPI worklist categorizes record pairs based on why they require human intervention. A search of the worklist cannot be initiated without selecting at least one category. A record pair may belong to more than one category simultaneously.

Review Category: The link weight of a record pair falls between the autolink and review thresholds. These are "potential links"—record pairs that have not been automatically linked and appear on the worklist for review because they have the highest link weight among non-linked pairs. Staff must use their judgment and the available tools (Comparison Detail, Whole Record Viewer) to determine if these records represent the same person and should be manually linked.

Validate Category: The link weight of a record pair falls between the validate and autolink thresholds. These are "links"—record pairs that have been automatically linked but appear on the worklist for validation because they have the lowest link weight among linked pairs. Staff should validate that the automatic linking decision was correct and can unlink if the system made an error.

Duplicate Category: Two records with a link weight above the autolink threshold belong to the same data domain (same facility/source). In EMPI, records from the same facility with different MRNs should NOT be automatically linked, even if they score high enough, because this indicates a duplicate within the source system. Duplicates should be sent back to the originating facility for MRN merge resolution. If the link weight is below the autolink threshold, the pair may appear in Review category instead (may not be a true duplicate).

Overlap Category: Two records are linked, but have different MPIIDs. This violates the EMPI rule that linked records must share the same MPIID. Overlaps often result from complex unlinking operations or MPIID reassignments. Resolution requires either linking the records and synchronizing MPIIDs (Same button) or unlinking them (Different button).

Overlay Category: Two records are not linked, but have the same MPIID. This violates the EMPI rule that records with the same MPIID must be linked. Overlays often result from unlinking duplicates without assigning new MPIIDs. Resolution requires either linking the records (Same button) or unlinking with MPIID reassignment (A is Different or B is Different button).

Open-chaining Category: A state in which two records are both linked to a third record but not linked to each other. For example:

• Records A and B are linked
• Records B and C are linked
• Records A and C are NOT linked

This creates a logical impossibility requiring human intervention. The Whole Record Viewer is particularly helpful for unraveling open-chaining issues.

Deterministic Category: A deterministic identifier could not be honored for a record pair. Deterministic identifiers are configured to force automatic linking or unlinking under certain circumstances (e.g., same SSN must link, different MRNs at same facility must not link). When a conflict occurs (e.g., two records have matching deterministic link identifier but also have matching deterministic unlink identifier), human intervention is required.

---

The search panel at the top of the worklist page allows users to filter which worklist items to display. Effective use of filters is essential for managing large worklists and focusing on specific types of issues.

Category Filter: Select one or more categories (Review, Validate, Duplicate, Overlap, Overlay, Open-chaining, Deterministic). The "All" checkbox selects all categories simultaneously. Note that selecting "All" may cause the worklist to display more slowly in larger deployments. At least one category must be selected to initiate a search.

Data Source Filter: The "Compare Data Sources for Pairs" section has two drop-down lists labeled "One Record" and "The Other Record". This allows sophisticated filtering:

• Display pairs where both records are from the same source: Set "One Record" to (any data source) and "The Other Record" to (same data source)
• Display pairs where records are from different sources: Set "One Record" to (any data source) and "The Other Record" to (different data source)
• Display pairs where one record is from a specific source: Choose the specific source from "One Record" and (any data source) from "The Other Record"
• Display pairs between two specific sources: Choose specific sources in both drop-downs

Data sources in the drop-down lists are populated from the Data Source List field under the General tab in the Definition Designer.

Time Created and Time Updated Filters: Specify date/time intervals during which records were created or updated. This allows focusing on recent activity (e.g., "show me all Review items created in the last 7 days") or historical investigation (e.g., "what duplicates were identified during the October data load?").

Link Reason Filter: Filter by the reason a pair was or was not linked. Options include:

• Threshold: Linked or unlinked due to link weight
• Transitivity: Two records linked because both are linked to a common third record
• Deterministic: Linked or unlinked due to a deterministic identifier
• Domain Conflict: Records from same data domain (facility) with different MRNs
• Manual: Manually linked or unlinked by a user
• Rule: Linked or not linked due to a user-implemented rule

Secondary Reason Filter: Occasionally, there is a reason for a pair to appear on the worklist that is different from or more specific than the link reason. For example, a user-defined rule may cause a pair to be linked, and the secondary reason field can capture which specific rule was triggered.

Minimum or Maximum Link Weight Filter: Specify the minimum and maximum link weights a record pair needs to appear in the worklist. This is useful for examining borderline cases (e.g., "show me all pairs with link weight between 22 and 26 to see what's happening near the autolink threshold of 24").

Custom Status Filter: Filter by custom status values that have been assigned to worklist items. Custom statuses are user-defined workflow tracking values (e.g., "Sent to Hospital", "Pending Review", "Escalated"). The filter provides both "Custom Status Includes" (show only items with this status) and "Custom Status Excludes" (hide items with this status).

ID Filter: Enter the internal ID that must belong to a record in order for its pair to be displayed. Clicking the "Search" button beside this field opens the InterSystems EMPI Record Search window, which allows searching by multiple criteria including MPIID, SSN, name, date of birth, and MRN. You can select multiple results to appear on the worklist.

Assignment Filter: Filter worklist items by assigned user. This field automatically displays the username of the current user by default, showing items assigned to you. You can also select (unassigned) to see only worklist items not assigned to any user, or select a different username to see items assigned to other users.

---

Each row in the worklist results table represents a pair of records (a worklist item). The table displays extensive information to help users understand the relationship between the records and decide on appropriate action.

Category: Displays the worklist category (Review, Validate, Duplicate, Overlap, Overlay, Open-chaining, Deterministic). A record pair may belong to more than one category, and all applicable categories are shown.

Link Status: Identifies a record pair as:

• Link: Records are linked and have the same MPIID
• Potential Link: Records are not linked but have a high link weight (between review and autolink thresholds)
• Non-Link: Records are not linked
• Conflict: Additional designation for pairs with overlay or overlap issues

Link Reason: Explains why the pair was or was not linked together. Displays one of: Threshold, Transitivity, Deterministic, Domain Conflict, Manual, or Rule. When a record pair satisfies more than one link reason, only the highest-priority reason is displayed (hierarchy: Manual > Rule > Deterministic > Domain Conflict > Transitivity > Threshold).

Secondary Reason: If configured, shows an additional reason for the linkage status, often used to capture which specific user-defined rule caused a linking decision.

Link Weight: Displays the numeric link weight score for the record pair. This is the sum of all parameter agreement weights and indicates the overall similarity between the two records. Higher scores indicate greater similarity.

Agreement Pattern: Displays a string where each character represents one weighted parameter in the matching configuration. For example, "XLNHX" for a configuration with 5 weighted parameters. The letters indicate:

• X (Negative): Agreement weight below zero—values are almost certainly not the same
• L (Low): Agreement weight above zero but less than half of the agreement threshold—values are probably not the same
• N (Neutral): Agreement weight is zero—no information
• H (High): Agreement weight is greater than half the agreement threshold—values are likely the same

Time Created: Displays the date and time when the worklist item was created (when the record pair first met the criteria for appearing on the worklist).

Time Updated: Displays the date and time when the worklist item was last modified (e.g., when link status changed, when custom status was assigned).

Alternate Views: Icons that can be clicked to display additional tools for analyzing the record pair:

• Comparison Detail icon: Compare record details side-by-side
• Record History icon: View the complete history of changes to a record
• Whole Record Viewer icon: View a record alongside all records linked to it
• Composite View icon: View the composite (golden) record
• Audit Log icon: View the audit history for a record

Unique Identifier: Displays an identifier that uniquely identifies each patient record, in the format :. For example, "PHLS:S2345" indicates a record from PHLS facility with MRN S2345.

Summary: Displays summary data for each patient as defined by the Record Summary Display Expression in the General tab of the Definition Designer. Typically shows name, date of birth, and other key identifiers. Clicking on a link in this column opens the Record History page for that record.

Record ID: The internal InterSystems EMPI record ID (not always displayed by default, but can be added via Column Selector).

MPIID: The Master Patient Index ID assigned to each record. Linked records should have the same MPIID; overlaps show different MPIIDs for linked records; overlays show the same MPIID for unlinked records.

Data Source: The facility or system from which each record originated.

Data Domain: The scope established by a particular site. Records from the same data domain (facility) with different MRNs create duplicates.

Pattern: See Agreement Pattern above.

Custom Status: Displays any custom status value assigned to the worklist item for workflow tracking.

Assignment: Displays the username of the user to whom the worklist item is assigned, if any.

---

Users can customize the worklist display to show information most relevant to their workflow.

Worklist Column Selector: Click the "Columns" button at the top of the worklist page to access the Worklist Column Selector pop-up. This interface allows selecting which columns should be displayed in the worklist table. Available columns include:

• Category, Link Status, Internal Link Status, Link Reason, Secondary Reason
• Link Weight, Agreement Pattern
• Time Created, Time Updated
• ID, Unique Identifier, Summary
• Data Source, Data Domain
• MPIID, Facility, MRN, Names, SSN, Gender, Birth Date
• Addresses, Telecoms (phone numbers)
• Assignment, Custom Status

Select desired columns and click "Apply", followed by "OK" to dismiss the window. Some columns are standard (same for all record pairs), and some are specific to the records being displayed (patient demographics).

Worklist Colors: Colors in the worklist provide visual cues about the degree of agreement or disagreement between records and fields. By default:

• Orange: Two values are not at all similar
• Yellow: Two values are somewhat similar
• Blue: Two values are very similar
• White: Values are neutral, data is missing, values did not contribute to link weight, or system data

Users can customize colors at the user level or system level by clicking the "Colors" button at the top of the page. Custom color schemes help users quickly identify which fields match, which conflict, and which are neutral.

Rows Per Page: Users can specify the number of rows to display on each page (e.g., 20, 50, 100). Smaller values provide faster page load times; larger values reduce pagination navigation.

Sort Order: Users can specify how the worklist is ordered, either by ascending or descending link weight. Ordering by descending weight shows the highest-scoring potential links first (most likely to be true matches). Ordering by ascending weight shows the lowest-scoring validated links first (most likely to be false matches requiring unlink).

Pagination Controls: Navigate between pages using the links labeled "First", "Previous", "Next", and "Last", or by entering a valid page number in the text box. For large numbers of results, some navigation options may be disabled to prevent long load times.

---

The worklist provides multiple ways to search for specific items or groups of items:

Filter-Based Search: The primary search method uses the search panel filters described earlier. Users select criteria (category, data source, date range, link weight range, custom status, assignment) and click "Refresh the Worklist" to update the display. Only record pairs matching ALL selected criteria are displayed.

Record Search Window: Clicking the "Search" button beside the ID field opens the InterSystems EMPI Record Search window. This allows searching for worklist items by patient identifiers:

• MPIID: Master Patient Index ID
• SSN: Social Security Number
• Name: Given name, family name, or full name
• Date of Birth: Full or partial date
• MRN: Medical Record Number
• Other identifiers: Depending on configuration

After entering search criteria, click "Search" to view results. Multiple results can be selected by checking the appropriate checkboxes. Click "OK" to close the window and display all selected records' worklist pairs on the worklist.

Comment Field Search Features: The Comment text box at the top of the worklist has two related features:

• Clear Comment: Once a comment has been applied to an action, it does not automatically disappear from the comment field. Users must manually click "Clear Comment" before performing more actions if they don't want the same comment applied.
• Previous Comment: Selecting the "Previous Comment" checkbox reuses the last comment that was applied to an action, streamlining repetitive workflow (e.g., "Sent to facility for MRN merge" can be quickly applied to multiple duplicates).

Last Search Filters: Clicking the "Last Search Filters" button repopulates the search panel with the filters that were used the last time the worklist was successfully generated. This allows users to quickly return to a previous search without manually re-entering all criteria.

Show Empty / Hide Empty: The search panel automatically hides filter fields that do not contain search criteria after a search is performed. Clicking "Show Empty" displays all filter fields; clicking "Hide Empty" hides empty fields again. This keeps the interface clean while still providing access to all filters.

Reset Filters: Clicking "Reset Filters" clears all search criteria and returns the search panel to its default state, allowing users to start a new search from scratch.

---

Examining record pairs near the autolink threshold is a critical quality assurance activity. Record pairs that score just above the autolink threshold are automatically linked but may be borderline cases. Record pairs that score just below the autolink threshold are not linked but may be true matches. The Threshold Adjuster tool helps identify and analyze these borderline cases.

Accessing the Threshold Adjuster: The Linkage Definition Threshold Adjuster is typically accessed through the Definition Designer or configuration menu. It provides a visual representation of how record pairs are distributed across link weight scores.

Histogram Display: The tool displays a bar chart (histogram) where:

• X-axis: Link weight scores (e.g., 0 to 40)
• Y-axis: Number of record pairs with each score
• Three vertical lines indicate: Review threshold (red), Autolink threshold (blue), Validate threshold (green)
• Bars show how many record pairs fall at each score level

Interpreting the Histogram: The distribution shape reveals important information:

• Large bars near the autolink threshold: Many borderline cases exist; these may require manual review or threshold adjustment
• Large bars below the review threshold: Many clear non-matches (good)
• Large bars above the validate threshold: Many high-confidence matches (good)
• Gaps or unusual patterns: May indicate data quality issues or matching configuration problems

Identifying Borderline Pairs for Review: Use the Link Weight filter in the worklist search panel to focus on borderline cases. For example, if the autolink threshold is 24:

• Set Minimum Link Weight to 22, Maximum to 26
• Select "Review" and "Validate" categories
• This displays all pairs scoring within ±2 points of the threshold

Review these borderline pairs using Comparison Detail or Whole Record Viewer to assess:

• Are the automatically-linked pairs (24-26) truly the same person?
• Are the non-linked pairs (22-23) truly different people?
• Should the threshold be adjusted based on these findings?

Threshold Tuning Process: Based on analysis of borderline cases: 1. Review a sample of pairs just above the autolink threshold (Validate category) 2. Review a sample of pairs just below the autolink threshold (Review category) 3. Calculate false positive rate (linked but shouldn't be) and false negative rate (not linked but should be) 4. Adjust threshold up (more conservative) if too many false positives 5. Adjust threshold down (more aggressive) if too many false negatives and false positive rate is acceptable 6. Re-run the Threshold Adjuster to visualize the impact of proposed changes 7. Implement changes and monitor results

Validate Category Priority: The Validate category specifically contains automatically-linked pairs that scored between the autolink and validate thresholds. These are the lowest-confidence automatic links and should be prioritized for quality assurance review. Regular validation sampling ensures that the autolink threshold is set appropriately and the matching configuration is performing well.

---

Sometimes a worklist search returns no results. Understanding why this happens helps users refine their search criteria and avoid frustration.

No Records Match Selected Criteria: The most common reason for empty results is simply that no worklist items currently exist that match all of the selected filter criteria. For example:

• Searching for "Duplicate" category items from a specific facility when no duplicates currently exist for that facility
• Searching for items with link weight above 30 when all high-scoring pairs have already been processed
• Searching for items assigned to a specific user when that user has no current assignments

Logically Impossible Filter Combinations: Some combinations of filters are mutually exclusive or logically impossible:

• Selecting "Overlap" category (linked records with different MPIIDs) and filtering by Data Source "One Record: PHLS, The Other Record: (same data source)" when no overlaps exist within PHLS
• Specifying Minimum Link Weight higher than Maximum Link Weight
• Filtering by a Custom Status that has never been assigned to any worklist items

Identifier Search Errors: When using the Record Search window to search by patient identifiers:

• Spelling errors: Searching for "John Smith" when the actual name is "Jon Smith" will return no results
• Formatting differences: Searching for SSN "123-45-6789" when stored format is "123456789"
• Partial matches not supported: Searching for full name when only first and last name fields are populated
• Case sensitivity: Some search fields may be case-sensitive

Time Range Excluding Results: The Time Created and Time Updated filters may exclude all results if:

• Date range is in the future
• Date range is before any worklist items existed (e.g., before EMPI implementation)
• Date range is too narrow (e.g., single day when looking for rare category like Open-chaining)

Data Source Filter Issues: The Compare Data Sources filter may return no results if:

• Selected data source does not exist in the Data Source List configuration
• Spelling of data source name doesn't match configuration exactly
• Data source has no records in EMPI
• Combination of "One Record" and "The Other Record" selections doesn't match any actual pairs

Category Selection Issues: The worklist search requires at least one category to be selected:

• If no categories are selected, clicking "Refresh the Worklist" will not execute the search
• If only rarely-occurring categories are selected (e.g., only "Deterministic"), results may be empty if no such conflicts currently exist

Troubleshooting Empty Results:

1. Simplify criteria: Start with minimal filters (e.g., just one category) and gradually add filters to narrow results 2. Check spelling and formatting: Verify identifier searches match data format exactly 3. Expand time ranges: Use broader date ranges or remove time filters entirely 4. Verify data source names: Check that selected data sources match configuration exactly 5. Review category selection: Ensure at least one category is selected and that category has items on the worklist 6. Use "Show Empty": Display all filter fields to verify all intended criteria are actually set 7. Try "Last Search Filters": Return to the most recent successful search to verify filter values 8. Check overall worklist size: Click "Refresh the Worklist" with minimal filters (e.g., just "All" categories) to see if ANY items exist

When No Results Are Expected: In some scenarios, empty results are actually the desired outcome:

• Searching for specific user assignment when clearing backlog (empty result = all items processed)
• Searching for specific Custom Status after workflow completion
• Searching for data sources after migration is complete
• Confirming that a category of issues (e.g., Overlaps) has been fully resolved

Understanding when empty results are expected versus unexpected helps users distinguish between successful worklist management and search configuration errors.

---

Effective worklist management requires understanding how to combine filters for specific workflow scenarios:

Scenario 1 - Morning Triage Workflow: A worklist specialist starts their shift and wants to see the highest-priority items requiring attention:

• Filters: Category = Review, Validate, Overlap, Overlay, Open-chaining; Assignment = or (unassigned); Order By = Descending link weight
• Rationale: Focus on conflict categories and borderline matches first, as these are most critical for data quality
• Expected Results: High-priority items requiring expert judgment

Scenario 2 - Duplicate Management: A data quality coordinator needs to send duplicates to source facilities for MRN merge:

• Filters: Category = Duplicate; Custom Status Excludes = "Sent to Hospital"; Order By = Data Source
• Rationale: Identify duplicates not yet sent to facilities, grouped by source for batch processing
• Actions: For each duplicate, assign Custom Status = "Sent to Hospital", add comment with date/contact info
• Expected Results: List of duplicates requiring facility notification, grouped for efficient processing

Scenario 3 - Quality Assurance Sampling: A supervisor wants to validate that the autolink threshold (24) is set appropriately:

• Filters: Category = Validate; Minimum Link Weight = 24; Maximum Link Weight = 26; Rows per page = 20
• Rationale: Sample the lowest-scoring automatic links (just above threshold) for quality review
• Actions: Review each pair using Comparison Detail; unlink if inappropriate; track false positive rate
• Expected Results: Sample of borderline automatic links for quality assessment

Scenario 4 - Backlog Reduction: A team is working to clear a backlog of Review items:

• Filters: Category = Review; Time Created = (older than 30 days); Assignment = (unassigned); Order By = Time Created ascending
• Rationale: Focus on oldest unassigned review items to reduce backlog age
• Actions: Assign items to team members; resolve using Link/Different buttons
• Expected Results: Oldest review items requiring resolution

Scenario 5 - Specialist Escalation: A lead identifies complex cases requiring specialist attention:

• Filters: Category = Overlap, Overlay, Open-chaining; Link Weight = (no filter); Assignment = (unassigned)
• Rationale: Identify linkage conflict categories requiring expert analysis
• Actions: Assign to specialist user; add comment explaining complexity
• Expected Results: Complex cases routed to appropriate expertise level

Scenario 6 - Facility-Specific Review: A facility liaison wants to review all worklist items involving their facility:

• Filters: Category = All; Data Source One Record = MEMORIAL_HOSP; Data Source The Other Record = (any data source)
• Rationale: See all worklist items where at least one record is from Memorial Hospital
• Expected Results: Complete view of Memorial Hospital's data quality issues

Scenario 7 - Post-Data-Load Validation: After a bulk data load from a new source, technical staff want to validate matching performance:

• Filters: Category = Review, Validate; Data Source One Record = NEW_CLINIC; Time Created = (last 7 days); Order By = Descending link weight
• Rationale: Review matches involving newly loaded source to ensure matching configuration performs well with this data
• Expected Results: Recent worklist items involving new source for validation

---

Advanced users develop sophisticated search strategies combining multiple features:

Multi-Step Search Refinement: 1. Start broad: Select all categories, no other filters 2. Note total count and distribution 3. Add filters incrementally: Category subset → Data source → Time range → Link weight 4. Observe how each filter reduces the result set 5. Refine to optimal working set size (e.g., 50-100 items)

Identifier Search + Filters: Combine the Record Search window with category filters for powerful targeted searches: 1. Click "Search" button, enter patient name "Smith" 2. Select all results with last name Smith (may be 20-30 patients) 3. Click OK to load all Smith worklist pairs 4. Apply Category filter = "Overlap, Overlay" to see only MPIID conflicts for Smith patients 5. Result: All MPIID conflicts involving anyone named Smith

Link Weight Band Analysis: Systematically analyze different scoring ranges:

• Band 1: Min=Review threshold, Max=Review+5 → Low-scoring potential links
• Band 2: Min=Autolink-5, Max=Autolink+5 → Borderline cases spanning threshold
• Band 3: Min=Validate-5, Max=Validate threshold → Low-confidence automatic links
• Review representative samples from each band to assess matching performance across score spectrum

Custom Status Workflow Tracking: Use custom status values to implement multi-stage workflows:

• Initial triage: Assign "Reviewed" status after first-level review
• Escalation: Assign "Escalated" status when sending to specialist
• External coordination: Assign "Sent to Hospital" when duplicate sent to facility
• Resolution tracking: Assign "Resolved" status after action taken
• Filter by status to see items at each workflow stage

Assignment-Based Team Management: Coordinate team worklist processing:

• Lead assigns batches of Review items to team members by name
• Each team member filters by Assignment = to see their queue
• Lead filters by Assignment = (unassigned) to see unallocated work
• Lead filters by Assignment = to check individual progress

Time-Based Analysis: Use Time Created and Time Updated to analyze worklist dynamics:

• Age analysis: Time Created > 90 days ago → Long-standing unresolved items
• Recent activity: Time Updated in last 24 hours → Recently modified items
• Inactivity: Time Created > 30 days AND Time Updated = Time Created → Stale items never touched
• Volume trends: Count by Time Created in weekly buckets → Worklist growth/reduction rate

Data Source Pattern Analysis: Identify systemic data quality issues by source:

• Run search for each data source individually
• Compare worklist item counts across sources
• Identify sources with disproportionate duplicate rates
• Investigate sources with unusual category distributions (e.g., many overlaps)
• Report findings to source system data quality teams

These advanced techniques help experienced users maintain worklist efficiency, identify systemic issues, and coordinate team-based resolution workflows.

---