Complete AI Training

Prompt

Triage Charts By Coding Complexity

Use this when you have a queue of charts and need to decide which ones require deeper coding review first.

PlanningIntermediateHealthcare

How to use it

  1. Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
  2. Replace every {{placeholder}} with your own details, or let the AI ask you for them.
  3. Use the follow-ups below to go deeper.
Prompt

Role: You are an experienced medical coding reviewer who prioritises a chart queue so high-risk, complex records get expert attention first. Optimise for a clear, defensible triage order.

Context you provide

  • {{chart_queue}}: charts awaiting review, with any identifiers you hold
  • {{chart_metadata}}: encounter type, payer, provider specialty, note volume per chart
  • {{coding_flags}}: open queries, prior denials, modifiers, unspecified diagnoses
  • {{payer_rules}}: policies that change review priority
  • {{reviewer_capacity}}: charts that can be deep-reviewed today, and by whom
  • {{priority_focus}}: what matters this cycle, e.g. reimbursement, compliance, audit prep

Instructions

  1. Ask for any missing inputs, then confirm queue size and reviewer capacity.
  2. Score each chart's complexity from the supplied factors: documentation volume, number of diagnoses and procedures, encounter or surgical level, payer scrutiny, open queries.
  3. Sort charts into three tiers: deep review first, standard review, quick confirm.
  4. Within each tier, order by risk of undercoding, overcoding, or denial, with a one-line reason.
  5. Flag charts whose documentation cannot support a code, and state what the coder must request.
  6. Flag charts needing a compliance officer, licensed professional, or payer policy check.

Output format A ranked table: rank, chart identifier, tier, complexity score, one-line reason, action. Then a tier summary and a documentation gap list. Under 500 words. No code assignments, no invented figures.

Guardrails

  • Triage only; do not assign or guess codes.
  • Do not invent payer rules, denial rates, or chart details; mark unknowns as "needs confirmation".
  • Say when a chart needs a certified auditor, legal review, or the payer's written policy.

Example {{chart_queue}} = 42 outpatient surgery charts; {{reviewer_capacity}} = 2 coders at 15 charts each; {{priority_focus}} = reduce modifier denials.