Workflow chains
AuraScore 81/100

Prior Authorization Clinical Evidence Synthesis Pipeline Design

Design an automated multi-agent workflow chain linking medical policy extraction, patient chart synthesis, and authorization packet generation.

Use this template when configuring an automated prior authorization clinical review chain between clinical EHR records and health plan coverage policies. It helps healthcare automation leads eliminate administrative delays while mitigating denial risks.

Template

Role: Lead Healthcare Interoperability and Clinical Automation Systems Specialist.

Context

  • Payer Policy Corpus: {{payer_guideline_corpus}}
  • Clinical Domain: {{clinical_specialty}}
  • Interoperability Interface: {{ehr_fhir_endpoint}}
  • Risk Profile: {{denial_risk_tolerance}}
  • Attestation Requirements: {{physician_attestation_step}}
  • Operational SLA: {{sla_turnaround_hours}}

Task

Author a comprehensive systems specification report for a multi-agent prior authorization pipeline that parses {{payer_guideline_corpus}}, extracts relevant patient clinical records via {{ehr_fhir_endpoint}}, detects clinical documentation gaps, and compiles an evidence packet for {{clinical_specialty}} within {{sla_turnaround_hours}}.

Method

  1. Define the policy parsing agent that decomposes unstructured coverage policies from {{payer_guideline_corpus}} into computable rule graphs.
  2. Design the EHR query synthesis agent that generates targeted FHIR queries against {{ehr_fhir_endpoint}} for required diagnostic proofs.
  3. Formulate the chart extraction and evidence matching agent that correlates retrieved labs, imaging, and clinical notes against policy prerequisites.
  4. Construct the gap detection agent that calculates coverage deficiency risks under {{denial_risk_tolerance}} settings.
  5. Detail the automated justification drafting agent that compiles medical necessity summaries referencing specific clinical dates and codes.
  6. Integrate the mandatory physician sign-off agent configured around {{physician_attestation_step}}.
  7. Architect the electronic submission and payload monitoring agent that tracks downstream payer receipt and status codes.
  8. Specify the circuit-breaker protocol when documentation gaps cannot satisfy payer necessity thresholds within {{sla_turnaround_hours}}.

Constraints

  • MUST incorporate {{physician_attestation_step}} prior to formal transmission to the payer.
  • MUST NOT submit authorization packets with unresolved gaps exceeding {{denial_risk_tolerance}}.
  • All EHR record queries MUST strictly query data via {{ehr_fhir_endpoint}} using least-privilege scoping.
  • Turnaround time for the complete synthesis cycle MUST conclude within {{sla_turnaround_hours}}.

Output format

Deliver a systems design report formatted as:

  1. Prior Authorization Agent Chain Topology
  2. Policy Ingestion & FHIR Evidence Retrieval Logic
  3. Gap Analysis & Denial Prevention Subsystem
  4. Clinician Attestation Gateway & Packet Generation
  5. Latency, Resilience, and SLA Governance Length must be between 850 and 1,250 words.

Self-review

  • Are all 6 variables referenced and operationalized in the pipeline steps?
  • Does the gap detection stage explicitly use {{denial_risk_tolerance}}?
  • Is the clinician attestation gate designed as an unskippable checkpoint?
AuraScore breakdown
81/100Provisional
Instruction clarity15/15 · Strong

Explicit role, a named task, and discrete steps the model can follow.

Context architecture12/12 · Strong

Background, inputs and variables the model needs before it starts.

Constraint engineering12/12 · Strong

Hard boundaries — what the model must and must not do.

Output specification6/14 · Thin

A named, field-level shape for the response.

Reasoning structure10/10 · Strong

Ordered work items that force analysis before an answer.

Model compatibility10/10 · Strong

Length and structure that travel across frontier models.

Token efficiency5/10 · Thin

Signal density — instruction weight without padding.

Reusability7/7 · Strong

Documented variables so the scaffold adapts to new inputs.

Robustness3/5 · Adequate

Quality bar, assumptions and behaviour when inputs are thin.

Observed performance1/5 · Thin

How much real usage the template has behind it.

ai-agents
agents-workflows
healthcare-life-sciences
prior-authorization
fhir
health-payer-operations