Audit Sentinel AI

Documentation & Help

How to Read Your Audit Report

A plain-language guide to every section of your Audit Sentinel analysis — what it means, why it matters, and exactly what to do with it.

Section 1 🏅 The Grade Card

This is the first thing you see after an audit completes. It gives you the bottom line in three elements: a numeric score, a letter grade, and a plain-language status label.

Score	Grade	What It Means
95–100	A	Excellent — your billing is fully supported by the documentation
90–94	A−	Very strong — minor opportunities to tighten documentation
80–89	B	Good — a few documentation gaps that could cause payer questions
70–79	C	Needs improvement — meaningful mismatches between codes and documentation
60–69	D	High risk — significant overcoding or undercoding detected
Below 60	F	Critical — substantial compliance exposure present

The status label (Compliant / Needs Improvement / High Risk) below the grade gives a quick plain-language read. The comparison summary explains in one or two sentences the most important factor that drove the score.

Tip: A score in the 80s is not a failing grade — it often means one specific documentation element is missing that, if added, would push you to 90+. Read the CDI section in the Ideal Analysis before changing your billing.

Section 2 🚨 Compliance Flag

If a red banner appears beneath the grade card, the AI has identified a potential overcoding situation — the code you submitted appears to exceed what the documentation can clinically support under the current CPT E/M guidelines (2023 revised standards).

This does not mean you've committed fraud. It means the documentation, as written, may not survive audit scrutiny at the billed level. Either the note should be strengthened before billing, or the code should be reviewed.

If no banner appears, your submitted code is at or below what the documentation supports — you're in the clear on the overcoding side.

Section 3 📊 Risk Profile

Five colored bars show where your exposure lies. Each bar represents the percentage of your total risk attributed to that category. The longest bar is your primary area of concern.

UndercodingYou submitted a lower code than the documentation supports — you may be leaving reimbursement on the table.

OvercodingYou submitted a higher code than the documentation supports — this is the highest audit risk category.

Modifier IssuesModifiers were missing, incorrect, or not justified by the documentation.

Diagnosis (ICD-10)Reported diagnoses don't match what the note addresses, or medical necessity isn't established.

OtherMiscellaneous issues including encounter type mismatches and time-based billing inconsistencies.

Tip: A high Undercoding bar is actually good news — it means you're billing conservatively and may be eligible for more reimbursement. Check the Ideal Analysis to see what the documentation would actually support.

Section 4 ✅ ❌ What You Got Right / Issues Found

Two side-by-side cards listing the specific findings from the audit.

✅ What You Got Right — Concrete things the AI confirmed are properly documented and support your billing. These are the elements you want to keep doing consistently in every note — not just praise, but a checklist of what's working.

❌ Issues Found — Specific gaps, mismatches, or missing elements. Each item is actionable — it tells you exactly what the documentation is missing or what the coding error is.

How to use this: Take the Issues Found list directly to whoever documents or codes the encounter. These are your priority fixes — specific, concrete, and ordered by importance.

Section 5 💡 Improvement Recommendations

Numbered, prioritized action items — the AI's specific suggestions for what to change or add. Unlike the Issues Found list (which describes what's wrong), the Recommendations list tells you what to do about it.

Recommendations typically fall into three types:

1. Documentation additions — specific phrases or elements to add to the clinical note to support the billed level.

2. Code corrections — changing the submitted CPT code, ICD-10 code, or modifier to better match the documentation.

3. Process changes — workflow or documentation practice changes to prevent the same issue in future encounters.

Tip: If you disagree with a recommendation, open the Ideal Analysis section and read the MDM card. It will show exactly what clinical elements the AI used to reach its conclusion — and what's missing to support the next level.

Section 6 🧮 Point Deduction Log

Shows the individual deductions that lowered your score from 100, broken into three columns:

Code — the specific element that was deducted (e.g., MDM-COMPLEXITY, ICD10-NECESSITY).

Reason — plain-language explanation of why points were removed.

Points — how many points this deduction cost you.

If your score seems lower than expected, start here — it will tell you exactly where the math went and which single fix would recover the most points.

Section 7 📋 Full E/M Ideal Analysis

Click the "Full E/M Ideal Analysis" bar at the bottom of your report to expand this section. This is the AI's complete independent analysis of what the ideal billing would look like for this encounter — before considering what you submitted. It is organized into five cards:

🏥 E/M Result & Reimbursement

The CPT code the documentation actually supports, with the clinical basis for that determination and typical reimbursement range.

🩺 ICD-10 & Medical Necessity

Diagnoses identified in the note mapped to ICD-10 codes, plus an assessment of whether medical necessity is established for the billed level.

📝 Medical Decision-Making

Full MDM breakdown — problems addressed, data reviewed, and risk level — showing exactly how the supportable E/M level was determined.

⚠️ Audit Risk Flags

Elements that would draw payer or OIG scrutiny. Shows green and says "None detected" when the encounter carries low audit risk as coded.

📋 Documentation Improvement (CDI)

Specific phrases and documentation elements the provider could add to future notes to strengthen support for the billed level — without changing any clinical facts.

Important: CDI suggestions apply to future notes for similar encounters. Do not retroactively alter a signed note — that creates legal risk. Use this card as a prospective template for documentation improvement.

🔄 What To Do After Your Audit

Score	Recommended Next Step
95+	File with confidence. Archive the report for your records.
85–94	Review the CDI card. Implement documentation suggestions going forward.
75–84	Address the Issues Found list before resubmission. Consider querying the provider for clarification.
60–74	Do not submit without review. Work through the Recommendations and Deduction Log with your coding team.
Below 60	Hold the claim. This encounter needs clinical documentation review and likely a code correction before any submission.

A note on AI limitations: Audit Sentinel is a powerful auditing aid — not a substitute for certified professional judgment. It applies the current CPT E/M guidelines (2023 revised standards, updated through the current code set) consistently, but does not have access to your full patient history, payer-specific LCD policies, or specialty-specific coding guidelines that may supersede general rules. When in doubt, escalate to a certified coder (CPC, CCS) or compliance officer. The audit report is a starting point for the conversation, not the final word.

→ Run an Audit