Get all your news in one place.
100's of premium titles.
One app.
Start reading
Medical Daily
Medical Daily
Cole Mercer

Blue Cross Ties $942 Million in Added Hospital Costs to AI Coding, but Hospitals Say Patients Are Sicker

A growing share of hospital patients are being coded as medically complex, and the Blue Cross Blue Shield Association says artificial intelligence tools are a major reason. In an analysis released Sept. 24, the insurer group reported that complex cases rose from 37% of inpatient claims for its member plans at the start of 2023 to 40% by the end of 2025, with no matching change in the care patients received.

The dispute is usually framed as a money fight between hospitals and insurers. But diagnosis codes do not appear only on bills. They become part of the records that follow patients, which raises a question most people rarely ask: Does my hospital record accurately describe how sick I was?

The hospital industry disputes the insurer group's reading. It says today's inpatients are genuinely older and sicker, and that better documentation captures conditions that were missed before. Neither side has released patient-level chart reviews that would settle the question.


An Insurer Analysis Points to Coding, Not Sicker Patients

According to the Blue Cross Blue Shield Association, rising coding intensity added an estimated $942 million in costs for Blue plans over 2024 and 2025, compared with a 2023 baseline. About $653 million of that came from secondary diagnoses that moved more than 55,000 claims into higher-paying categories, about $11,000 per extra-complex case, Fierce Healthcare reported.

Secondary diagnoses are conditions listed alongside the main reason for a hospital stay. The association said AI tools can find them by scanning existing records or through ambient scribes, which listen during visits and draft clinical notes.

The group's central argument is a mismatch between diagnosis and treatment. Among major bowel procedures, claims coded at the highest complexity level rose from 10.2% to 22.7%. At the hospitals most likely to diagnose anemia caused by blood loss, patients with that diagnosis received blood transfusions less often than patients at other hospitals, 16.9% compared with 19.3%.

"The disconnect between diagnoses and treatment suggests that AI is identifying more billable conditions, not sicker patients," said Luke Chalker, the association's senior vice president of product and data science. Dr. Razia Hashmi, its vice president of clinical affairs, was more measured. "There may be an element of correct coding there, but the likelihood that this is technology-enabled upcoding is higher, in my view," she told reporters.


Hospitals Say Today's Inpatients Really Are Sicker

Readers should weigh the source. The analysis was produced by an insurer trade group with a financial stake in lowering hospital payments, and it was released as a white paper rather than a peer-reviewed study. It relies on claims data rather than medical charts, and the association acknowledged that charts would be a more direct test of whether patients were truly sicker.

The American Hospital Association argued in a fact sheet first published in July that an aging population and rising chronic disease are increasing patient complexity. It cited an AHA and Vizient analysis finding that hospital case-mix index, a standard measure of how sick patients are, rose about 5% from 2019 to 2024.

The hospital group also pointed to documented upcoding by insurers, including a 2025 finding by the Medicare Payment Advisory Commission that upcoding contributed to $40 billion in overpayments to Medicare Advantage plans. Hospital leaders have said AI helps them document care accurately and respond to claim denials.

Both things can be partly true. Some added diagnoses may reflect real conditions that were previously overlooked, while others may be thinly supported. The analysis cannot tell an individual patient which applies to them.


Your Chart, Your Bill, and the Right to Ask Questions

For patients, the most concrete issue is accuracy. A secondary diagnosis coded during a hospital stay can appear on insurance statements and in records shared with future doctors. If a condition was recorded but never discussed or treated, it may confuse later care decisions.

Patients do not need to assume anything is wrong. But after a hospital stay, especially a surgical one, it is reasonable to compare the insurer's explanation of benefits with the discharge summary and ask about any unfamiliar diagnosis. A doctor or the hospital's billing office can explain why a condition was listed.

Federal privacy law gives patients tools. Under HIPAA, people can request copies of their medical and billing records and ask providers to correct information they believe is wrong, according to the U.S. Department of Health and Human Services. Providers do not have to agree, but patients can generally submit a written statement of disagreement if a request is denied.

People with new or worsening symptoms should not wait on a billing review. Any clinical concern, including a diagnosis a patient does not understand, is best raised directly with a treating clinician.


Premiums, Open Enrollment, and the Next Round of Data

The association says higher coded costs lead to higher premiums and out-of-pocket costs for members, employers, and taxpayers, though it did not estimate the effect on any individual plan. An earlier association analysis of AI-enabled hospital billing, released in March, focused on anemia diagnoses after childbirth. It estimated that about $2.3 billion in inpatient and outpatient spending nationwide may be tied to AI-enabled coding.

The new report focuses on hospital inpatient care. Chalker said the association plans more analyses covering outpatient care and other diagnosis categories.

Fall open enrollment is a practical time for families to review plan costs and how their insurer handles billing disputes. Until independent chart reviews compare coded diagnoses with actual care, patients are best served by reading their own records and asking questions when something does not match their experience.


Key Questions Answered

What did the Blue Cross analysis find?

Complex inpatient cases rose from 37% to 40% of Blue plan claims between early 2023 and late 2025, adding an estimated $942 million in costs without a matching change in treatment.

What is upcoding?

It refers to billing for a more severe or complex condition than the care delivered supports. The insurer group says AI tools are contributing, and hospitals dispute that interpretation.

Do hospitals agree?

No. The American Hospital Association says patients are older and sicker and that documentation has improved, citing a roughly 5% rise in case-mix index from 2019 to 2024.

Is the analysis independent?

No. It was produced by an insurer trade group, relies on claims rather than charts, and was released as a white paper rather than a peer-reviewed study.

How can I check my own records?

Compare your insurer's explanation of benefits with your discharge summary, and request copies of your records if something looks unfamiliar.

Can I fix an error in my record?

HIPAA lets you request a correction. The provider may decline, but you can generally submit a statement of disagreement for your file.

Published by Medicaldaily.com

Sign up to read this article
Read news from 100's of titles, curated specifically for you.
Already a member? Sign in here
Related Stories
Top stories on inkl right now
Our Picks
Fourteen days free
Download the app
One app. One membership.
100+ trusted global sources.