Skip to main content

CDI: Generating Physician Queries Automatically

Duration: 55 min · Level: Advanced · Module: 3. AI Medical Coding & CDI · Focus: CDI, physician-query, documentation-improvement, revenue-integrity

Learning objectives

By the end of this lesson you will be able to explain and apply:

  • CDI query triggers
  • AHIMA/ACDIS query practice guidelines
  • AI query generation
  • Integration point
  • Query response tracking

Why this matters​

Clinical Documentation Improvement (CDI) is the practice of querying physicians to clarify documentation that is vague, conflicting, or insufficient for accurate coding.

Overview​

Clinical Documentation Improvement (CDI) is the practice of querying physicians to clarify documentation that is vague, conflicting, or insufficient for accurate coding. Traditionally CDI specialists manually review records; AI can generate query opportunities automatically from discharge documentation and route them in real-time during the hospitalization.

Key concepts​

Key idea

CDI query triggers: conflicting documentation (body of note says "sepsis" but diagnosis list says "infection"), clinical indicators without diagnosis (high WBC + antibiotics + fever but no sepsis diagnosis), vague diagnoses ("respiratory failure" vs "acute hypoxic respiratory failure")

  • AHIMA/ACDIS query practice guidelines: queries must be unbiased (present options without leading); must be based on clinical documentation not assumption; compliant queries ask for clarification, not new information
  • AI query generation: LLM reads the complete record, identifies clinical indicators (lab values, vitals, medications, imaging) that suggest diagnoses not explicitly documented, generates a compliant query text presenting the clinical evidence and asking the physician to clarify
  • Integration point: real-time CDI query generated when attending physician opens the patient's note for a new progress note entry → query displayed as EHR Best Practice Advisory → physician responds in EHR → coding agent receives response immediately
  • Query response tracking: track query response rate (target >85%), query agree rate (% physician agrees with suggested documentation), revenue impact per query ($X DRG improvement); this data feeds the AI model improvement loop
  • Compliance guardrails: AI-generated queries must be reviewed by a human CDI specialist or coded by a human before being sent to physicians in most health systems; full automation is a future state pending more validation data

Check your understanding​

Try to recall each answer before expanding it.

Q1. What do you know about CDI query triggers?

conflicting documentation (body of note says "sepsis" but diagnosis list says "infection"), clinical indicators without diagnosis (high WBC + antibiotics + fever but no sepsis diagnosis), vague diagnoses ("respiratory failure" vs "acute hypoxic respiratory failure")

Q2. What do you know about AHIMA/ACDIS query practice guidelines?

queries must be unbiased (present options without leading); must be based on clinical documentation not assumption; compliant queries ask for clarification, not new information

Q3. What do you know about AI query generation?

LLM reads the complete record, identifies clinical indicators (lab values, vitals, medications, imaging) that suggest diagnoses not explicitly documented, generates a compliant query text presenting the clinical evidence and asking the physician to clarify

Q4. What do you know about Integration point?

real-time CDI query generated when attending physician opens the patient's note for a new progress note entry → query displayed as EHR Best Practice Advisory → physician responds in EHR → coding agent receives response immediately

Q5. What do you know about Query response tracking?

track query response rate (target >85%), query agree rate (% physician agrees with suggested documentation), revenue impact per query ($X DRG improvement); this data feeds the AI model improvement loop

← Previous: H3.1 LLM-Based Diagnosis Code Extraction: The Architecture · Next: H3.3 CPT Coding for Outpatient & E&M Automation →

Part of Module 3: AI Medical Coding & CDI.