Top 10 Claim Denial Reasons and How to Fix Them [2026 Data] | Trustro.ai

U.S. healthcare providers lose billion annually to claim denials. Here are the 10 most common denial reasons with their CARC codes, prevention strategies, and how AI billing agents catch errors before submission.

The American Medical Association estimates that U.S. healthcare providers lose billion annually to claim denials. The average denial rate across specialties is 10% to 15% of submitted claims, and 65% of denied claims are never reworked or resubmitted (source: MGMA DataDive, 2024). That means the majority of lost revenue stays lost.

Most denials are preventable. The top 10 reasons account for over 80% of all denials, and each one has a specific prevention strategy. This guide covers each denial reason with its Claim Adjustment Reason Code (CARC), explains why it happens, and shows how to prevent it.

1. Missing or invalid patient information (CARC 16)

CARC 16: "Claim/service lacks information or has submission/billing error." This is the single most common denial reason, accounting for approximately 25% of all denials. It includes misspelled names, incorrect date of birth, wrong insurance ID numbers, and mismatched subscriber information.

Prevention: Run eligibility verification before every visit. Verify patient demographics against the payer's database at check-in. Automated eligibility checks catch 90% of these errors before the claim is ever submitted.

2. Duplicate claim submission (CARC 18)

CARC 18: "Exact duplicate claim/service." This happens when the same claim is submitted twice, often because staff resubmit after not seeing a response within the expected timeframe. It also occurs when multiple billers work the same account.

Prevention: Use a claims dashboard that shows submission status in real time. Track claim lifecycle from Draft to Ready to Submitted to Pending to Paid/Denied. A unified billing workflow prevents multiple people from submitting the same claim.

3. Service not covered by payer (CARC 50)

CARC 50: "These are non-covered services because this is not deemed a medical necessity by the payer." The payer determines the service was not medically necessary based on the diagnosis codes submitted, or the specific service is excluded from the patient's plan.

Prevention: Match ICD-10 codes to CPT codes according to payer-specific medical necessity guidelines. An AI billing agent can cross-reference the diagnosis with the procedure and flag mismatches before submission. For example, a payer may deny 97110 (therapeutic exercise) if the diagnosis is Z00.00 (general adult examination) rather than M54.5 (low back pain).

4. Authorization required (CARC 197)

CARC 197: "Precertification/authorization/notification absent." Many payers require prior authorization for specific procedures, imaging, or specialist referrals. If the authorization number is not included on the claim, it is denied automatically.

Prevention: Maintain a prior authorization tracking system. Flag appointment types that typically require authorization. Include the authorization number on the claim at the time of submission, not as an afterthought.

5. Timely filing limit exceeded (CARC 29)

CARC 29: "The time limit for filing has expired." Each payer has a submission deadline: Medicare is 365 days, most commercial payers are 90 to 180 days, and Medicaid varies by state. Once the deadline passes, the claim cannot be submitted regardless of its validity.

Prevention: Submit claims within 48 hours of the encounter. Automated claim generation from signed notes eliminates the lag between documentation and submission. Track aging claims and prioritize those approaching filing deadlines.

6. Incorrect procedure code (CARC 4)

CARC 4: "The procedure code is inconsistent with the modifier used or a required modifier is missing." Common examples: billing 99214 without the required modifier 25 when also billing a procedure on the same date, or using an incorrect place of service code.

Prevention: AI coding tools validate code combinations before submission. They check modifier requirements, bundling rules, and place of service consistency. For example, the system flags if you bill 99214 and 20610 (joint injection) on the same date without modifier 25 on the E/M code.

7. Patient not eligible on date of service (CARC 27)

CARC 27: "Expenses incurred after coverage terminated." The patient's insurance was not active on the date of service. This is common with patients who changed jobs, aged out of a parent's plan, or had a lapse in coverage.

Prevention: Real-time eligibility verification at check-in. Run the verification on the day of the visit, not when the appointment was booked. Insurance status can change between booking and the visit date.

8. Bundling and unbundling errors (CARC 97)

CARC 97: "The benefit for this service is included in the payment/allowance for another service/procedure that has already been adjudicated." Payers have National Correct Coding Initiative (NCCI) edits that bundle certain procedures together. Billing them separately results in denial.

Prevention: Check NCCI edits before submission. For physical therapy, this commonly affects 97110 and 97530 when billed together without appropriate documentation of distinct services. AI billing agents run NCCI edit checks automatically.

9. Coordination of benefits issue (CARC 22)

CARC 22: "This care may be covered by another payer per coordination of benefits." The patient has multiple insurance plans, and the claim was sent to the wrong payer (secondary instead of primary, or vice versa).

Prevention: Collect complete insurance information during intake, including any secondary coverage. Verify coordination of benefits order. Submit to the primary payer first, then submit the balance to the secondary payer with the primary's EOB attached.

10. Invalid diagnosis code (CARC 167)

CARC 167: "This (these) diagnosis(es) is (are) not covered." The ICD-10 code submitted is either invalid (deleted or outdated), does not support the procedure, or lacks the required specificity. For example, using E11 (type 2 diabetes) instead of E11.9 (type 2 diabetes without complications) or E11.65 (type 2 diabetes with hyperglycemia).

Prevention: Use current ICD-10 code sets (updated annually in October). AI coding tools suggest the most specific code supported by the documentation, reducing unspecified code usage by 60% to 80%.

How AI billing agents prevent denials

An AI billing agent runs every claim through a series of checks before submission: patient eligibility verification, NCCI edit validation, modifier requirement checks, diagnosis-to-procedure medical necessity matching, timely filing deadline tracking, and payer-specific rule compliance. Claims that fail any check are flagged for review rather than submitted with errors.

When denials do occur, the AI reads the CARC and RARC codes on the remittance advice, identifies the specific error, recommends the correction, and can resubmit the corrected claim automatically. Practices using AI billing agents reduce their denial rate by 35% to 45% and cut days in accounts receivable by 40% (source: Trustro platform data, Q1 2026).

Frequently asked questions

What percentage of claim denials are preventable?

Approximately 86% of claim denials are preventable according to the Healthcare Financial Management Association (HFMA). The most common preventable causes are eligibility verification failures, coding errors, and missing authorization numbers, all of which can be caught before submission.

How much does the average denial cost to rework?

The average cost to rework a denied claim is to depending on the complexity and the number of touches required (source: MGMA, 2024). For high-volume practices, this adds up to tens of thousands of dollars annually in administrative costs alone.

What is a good claim denial rate?

A denial rate below 5% is considered excellent. The industry average is 10% to 15%. Practices using AI billing agents with pre-submission claim scrubbing typically achieve denial rates of 3% to 5%.

Should I appeal denied claims?

Yes. Over 60% of denied claims are overturned on appeal when the correct information is submitted. The key is identifying the specific denial reason (CARC code), correcting the error, and resubmitting within the payer's appeal window (typically 30 to 90 days).

How quickly should I follow up on denied claims?

Follow up within 7 days of receiving the denial. Payers have appeal deadlines, and delays reduce your chance of recovery. AI billing agents that auto-identify and auto-appeal denials eliminate this follow-up lag entirely.

The bottom line

Claim denials are not random. They follow predictable patterns, and 86% are preventable. The top 10 reasons in this guide account for over 80% of all denials. Each one has a specific CARC code, a known cause, and a prevention strategy that can be automated.

For practices that want to stop losing revenue to preventable denials, the first step is pre-submission claim scrubbing. Trustro's AI Billing Agent runs every claim through eligibility verification, NCCI edits, modifier validation, and payer-specific rules before submission. See it work on your claims: /demo

Top 10 Claim Denial Reasons and How to Prevent Them