2026 Research Report

2026 Medical Billing Errors Report

BillKarma analyzed 47,000+ hospital bills submitted in 2025–2026. Here's what we found.

1 in 3 bills contain an error
$1,312 avg overcharge per bill
34% overall error rate
76% of disputes won

Key Findings

Five takeaways that every patient should know.

🫀
41%

Cardiac bills have the highest error rate

Cardiovascular procedures are misbilled at 41%, nearly double the national average. Complex coding and high procedure volume make this specialty especially prone to errors.

📋
28%

Duplicate charges are the most common error

More than one in four billing errors involve the same service billed multiple times. This rate is up 18% from 2024, driven by EHR copy-paste workflows.

📈
2.3×

Uninsured patients are overcharged far more often

Patients without insurance are 2.3 times more likely to receive an inflated bill. Without an insurer to push back, errors frequently go uncontested.

💰
52%

Large bills are especially error-prone

Bills exceeding $10,000 contain errors at a 52% rate — more than half. Higher complexity and longer hospital stays multiply opportunities for coding mistakes.

76%

Disputing works — three in four patients win a reduction

When patients formally dispute a bill, 76% receive a correction or reduction. The biggest barrier is knowing how and when to push back.

Error Rate by Specialty

Cardiac and emergency medicine top the list. Mental health and primary care have the lowest rates.

Types of Billing Errors

Duplicate charges and upcoding alone account for more than half of all errors.

  • 28% Duplicate charges
    Same service billed multiple times
  • 25% Upcoding
    Charged for more expensive service than performed
  • 18% Unbundling
    Separate charges for services that should be bundled
  • 14% Services not rendered
    Charged for something that didn't happen
  • 9% Incorrect patient data
    Wrong name/DOB/insurance causes billing errors
  • 6% Other coding errors
    Miscellaneous ICD-10/CPT mismatches

Error Rate Trend 2020–2026

After dipping in 2023, error rates have climbed to a six-year high.

Error rates spiked post-pandemic as staffing shortages increased coder workloads. The 2026 figure of 34% is the highest recorded in our dataset.

State Highlights

Geography matters. Where you live affects how likely you are to be overcharged — and how easy it is to fight back.

🏆

California

Lowest average resolution time — just 23 days when patients dispute a bill. Strong consumer protection laws make providers respond quickly.

⚠️

Texas & Florida

Highest average overcharge at $1,890 and $1,720 respectively. Both states have limited price transparency enforcement and high out-of-network billing rates.

💪

Illinois

Highest dispute win rate at 84%, aided by some of the strongest charity care laws in the nation and mandatory financial assistance programs.

What You Can Do

You don't need a lawyer or a billing expert. Four steps that work.

1

Request your itemized bill — not just the summary statement. Hospitals are required to provide one. This is where errors hide.

2

Compare each charge to Medicare reference rates using BillKarma. Any charge more than 3x the Medicare rate is worth disputing.

3

File a dispute letter within 30 days of receiving your bill. Cite the specific line items and reference rates. Written disputes get faster results.

4

If denied, escalate to your state insurance commissioner or attorney general's consumer protection office. Most disputes resolve at this stage.

Found an error in your bill? BillKarma helps you dispute it for free.

Fight Your Bill Now

Methodology

Sample size
47,312 bills processed January 2025 – March 2026
Method
Compared against Medicare Reference Rates, CMS NCCI edits, and AMA CPT code guidelines
Error definition
A charge ≥15% above the applicable reference rate, or a clear coding violation per NCCI or AMA guidelines
Geographic coverage
All 50 states; largest sample concentrations in CA, TX, NY, FL, and IL

Disclosure: BillKarma does not sell user data. All individual bills are fully anonymized before inclusion in aggregate analysis.