Task Management

Every denial, every prior-authorization decision, every RxRenewal, every co-sign request, and every payer follow-up call lands as a FHIR Task in a unified work queue. Tasks carry SLA deadlines; overdue items auto-escalate. An AI next-best-action engine suggests the optimal step. Skills-based routing sends the right task to the right person. Configurable views let each staff member see their world through the right lens. No more sticky notes, spreadsheets, or tribal knowledge — every follow-up action has a record, an owner, and a clock.

Key Capabilities

Skills-Based Task Routing

Tasks are routed to the team member whose skill tags match the task type and payer. Workers' comp denials go to the biller tagged with WorkersComp; complex appeals go to senior billers. Nurse tasks land on the nurse's queue; physician co-sign requests go to the supervising doctor. The routing engine evaluates up to 50 skill tags per user and 100 routing rules per org, assigning to the best-match user with the lowest active task count. If no skill match is found, the task drops to the default queue.

Configurable Queue Views

The unified queue surface displays all work items — denials, follow-ups, co-signs, renewals, referral authorizations, PA decisions — in a single view filtered by role, task type, priority, SLA status, and assignee. Users save and switch between named views ("My Tasks," "Denial Queue," "Overdue," "Co-sign Requests," "My Specialties"). Views are org-scoped and user-scoped. The queue loads in ≤2 seconds for up to 500 active tasks per org.

Escalation Timers

Every task carries an SLA deadline stored in SlaDeadlineDateTimeUTC. The SLA monitor runs every 15 minutes and auto-escalates tasks that exceed their deadline while in an active state. The default threshold is 30 days, configurable per task type and per org. Escalation creates a TaskStatusHistory row with status Escalated, reassigns the task to the practice manager, and sends an SMS/email alert. Overdue appeals generate an escalation task assigned to the org admin. All escalation events are recorded for audit.

Task Templates (Macros)

Pre-built and custom macros auto-populate action sequences: appeal letter templates with patient data, diagnosis codes, clinical rationale, and payer coverage criteria; standard follow-up sequences; payer call scripts. Selecting a macro executes the ActionSequenceText — populating templates, attaching documents, creating sub-tasks, and transitioning status — in one click instead of reconstructing the steps each time. Macro-driven letters reduce drafting time from 20 minutes to under 3 minutes. Target: ≥70% of appeal letters generated via macro.

Audit Trail on Every Task Action

Every action on a task is time-stamped, duration-logged, and actor-attributed. Status transitions are captured in TaskStatusHistory with the changer's UserID and timestamp. Time entries in TaskTimeLog record the action, duration, and user. Payer interactions (voice via Twilio, SMS, email) are auto-logged with caller ID, duration, transcript summary, and outcome. The audit trail viewer is searchable by patient, payer, task type, and date range — supporting compliance audits and payer disputes with zero gaps.

AI Next-Best-Action

For each active task, the AI engine suggests the optimal next step based on the task type, denial reason code, payer history, and elapsed time. Suggestions include "File appeal with medical records," "Request peer-to-peer review," "Call payer," "Collect medical records." The biller accepts, modifies, or dismisses; the action is logged. Target: ≥60% AI suggestion acceptance rate. Suggestions are pre-computed for the top 100 tasks in each user's queue and cached for 10 minutes, generated in ≤500ms p95 per task.

Payer Follow-Up Logging

Native voice (Twilio), SMS, and email follow-up logging to payers. Outbound calls are auto-logged with caller ID, duration, and a post-call outcome prompt. SMS and email interactions are captured with send/receive timestamps and content. Inbound responses are matched to the originating task by payer contact and task reference. TCPA consent is verified before any outbound call or SMS. Target: 100% of payer interactions logged on the task.

Task Board View

A Kanban-style board renders the unified work queue as columns representing task lifecycle stages: To Do → In Progress → Review → Done. Each column shows a count badge and is populated with task cards. Cards display: title (denial reason, Rx name, referral target), assignee avatar and name, priority badge (Urgent / High / Routine / Low), due date with SLA countdown (overdue in red, due-soon in amber, on-track in green), and task type tag (Denial, Prior Auth, RxRenewal, Co-sign, Referral, Follow-up).

Staff drag cards between columns to update status — dragging a card from "To Do" to "In Progress" transitions the FHIR Task status from Requested to In Progress and creates a TaskStatusHistory row. The board supports filtering by persona (biller-only, nurse-only, provider-only), priority, and source module (RCM, eRx, Referrals). Saved filter configurations persist per user. AI next-best-action suggestions appear inline on cards in the "To Do" column.

Persona Connections

• Nurse — Handles follow-up tasks: RxRenewal processing under standing orders, referral authorization follow-ups, medical record collection for appeals. Uses the queue's configurable views to filter by "Nurse tasks only." Processes tasks at level 40 — can update status and log time.
• Doctor — Receives escalated items, co-sign requests, and prior-authorization decisions in the unified queue. Signs off on tasks with one click from the chart or the queue. Benefits most from AI next-best-action suggestions on clinical follow-ups. Operates at level 60 — can re-assign and override.
• MA — Processes rooming tasks: document upload requests, intake completion follow-ups, and standing-order renewals. Uses filtered views for "MA tasks only." Operates at level 40.
• RCM / Priya — Owns the denial queue. Monitors SLA compliance, auto-escalations, and payer follow-up logs. Uses macros for appeal letter generation. Reviews AI next-best-action suggestions. Benefits most from the analytics dashboard and audit trail viewer. Operates at level 50–70 depending on role (biller vs. senior biller vs. practice manager).

Technical Highlights

• Task priority engine. Every WorkTask carries a PriorityString (Urgent / High / Routine / Low) that maps directly to FHIR Task.priority. The priority drives sort order in the queue and escalation thresholds. Urgent tasks escalate after 7 days; routine tasks after 30 days. Priority is assigned at task creation based on the source module's classification and can be overridden by the practice manager.
• Skill-tag matching. Users carry configurable skill tags (e.g., WorkersComp, ComplexAppeals, BlueCross). The routing engine matches WorkTask.TaskTypeString and SourceModuleString against assignee skill tags. Routing rules are configurable per org by the practice manager (level ≥70). Only senior billers (≥55) and practice managers can configure routing rules. Target: ≥85% routed tasks accepted by assignee.
• SLA timers. The SlaDeadlineDateTimeUTC field on every WorkTask drives the SLA monitor, which runs every 15 minutes as a scheduled job. Auto-escalation latency from deadline breach to reassignment is ≤5 minutes. Escalation creates TaskStatusHistory rows with ChangedByUserID set to the system user (indicating auto-escalation). The SLA monitor processes ≥10,000 active tasks per org per cycle.
• FHIR Task resource alignment. WorkTask is modeled as a FHIR R4 Task resource for interoperability with payer systems via the CMS-0057-F Provider Access API. The mapping covers Task.status, Task.priority, Task.code, Task.focus, Task.owner, Task.for, and Task.executionPeriod.end. TaskStatusHistory maps to Task.statusHistory. The FHIR Task endpoint is exposed for CMS-0057-F consumer use cases.
• Org-scoped isolation. All four Task Management noun-apps (WorkTask, TaskStatusHistory, TaskMacro, TaskTimeLog) are org-scoped with OrganizationID partition-key isolation. A denial at Practice A has no visibility to Practice B, even for the same patient. The patient's clinical facts remain global; the task references them via FK but does not own them.

Task Lifecycle Pipeline

The spine of this module ensures that no follow-up action falls through the cracks — every event that requires human attention becomes a FHIR Task with an owner, a clock, and a full audit trail:

1. Source module event (denial posted, RxRenewal received, co-sign requested) → WorkTask auto-created with type, priority, SLA deadline, and source reference.
2. Skills-based routing engine evaluates task against user skill tags → assigned to best-match user with lowest active task count.
3. AI next-best-action engine generates suggestion → displayed inline in queue and on task detail view.
4. User acts: accepts suggestion, applies macro, makes payer call, updates status → TaskStatusHistory row created, TaskTimeLog entry recorded.
5. SLA monitor polls every 15 minutes → overdue tasks auto-escalate to practice manager → SMS/email alert sent.
6. Task completed or cancelled → status finalized, cumulative time and action history preserved in audit trail.

Delivery Phases

Success Metrics

• Denial-to-task conversion: 100% of denials create a FHIR Task within 1 minute of 835 posting — unworked denials are silent revenue loss.
• SLA compliance rate: ≥95% of tasks resolved within their SLA deadline — exceeded SLAs mean missed appeal windows and delayed revenue.
• Auto-escalation accuracy: ≥90% of auto-escalations accepted (not dismissed) by the practice manager — high dismissal rates signal over-sensitive rules.
• AI next-best-action adoption: ≥60% of suggestions accepted by the biller — below 40% suggests the model needs retraining.
• Payer follow-up logging coverage: 100% of outbound payer calls/emails/SMS logged on the task — undocumented interactions are a compliance gap.
• Macro-driven appeal rate: ≥70% of appeal letters generated using macros — faster, more consistent, with required clinical data included automatically.
• Skills-based routing accuracy: ≥85% of routed tasks accepted by assignee (not reassigned) — high reassignment rates indicate misconfigured routing logic.

Module Dependencies

• Upstream — RCM creates tasks from denied claims, claim status updates, and appeal deadlines. eRx / EPCS creates tasks from RxRenewal, RxChange, and EPCS co-sign requests. Referrals creates tasks from authorization requests and specialist follow-ups. Eligibility creates tasks for prior-authorization requirements.
• Lateral — Coding / CDS provides clinical documentation referenced in appeal letter templates. Patient Portal surfaces patient-facing tasks (appointment reminders, document uploads, balance inquiries) that sync with the internal queue.
• Downstream — Payer Optimization consumes task resolution data for the ROI dashboard and payer performance tracking.

Try This in the Demo

• Nurse — Task Management screen →