
And in Shakespeare's play Hamlet, two characters, Rosenstern and Guildenstern, are sent off to their death by Hamlet.

And at the start of the play, there's a succession of coin tosses that produces 92 heads in a row.

Rosencrantz keeps winning and collecting his winnings, and Guildenstern becomes increasingly troubled by a world in which the same results occur again and again.

If artificial intelligence helps a coder to find information already supported by the record, the resulting may increase reimbursement for the hospital and the hospital should have received all along.

Payers have used data analytics and automated files for years to identify claim denials, payment reductions, and recovery.


Optum describes an advanced analytics for AI for claim review, coding checks, edits, and prevention of overpayments.

Cigna's provider instructions expressly address claims denied because of coding edits.

They also establish that the technology scrutinizing a claim for financial consequences is hardly a new development.

Calling one system data analytics and the other AI does not determine whether its conclusions are correct or incorrect.

Traditional edits and generative AI are a different technology, but both require accountable use.

The suggestion that a second code is inherently suspicious in billing overlooks the structure of the ICD-10 instructions, such as code first and use additional code can require multiple codes and specify their order.

A code also, note, indicates that two codes may be necessary without itself directing the sequence.

For hypertensive chronic kidney disease, for instance, an I-12 category is accompanied by an N18 code identifying the kidney stage.

Similarly, hypertensive heart disease with heart failure requires an additional I-50 category code identifying the heart failure type.

An AI reminder to check for missing companion codes may improve compliance.

These instructions predate AI and conventional coding software can flag them too.

AI can add value by helping locate supporting documentation scattered across a lengthy chart.

And in Shakespeare's play Hamlet, two characters, Rosenstern and Guildenstern, are sent off to their death by Hamlet.

And at the start of the play, there's a succession of coin tosses that produces 92 heads in a row.

Rosencrantz keeps winning and collecting his winnings, and Guildenstern becomes increasingly troubled by a world in which the same results occur again and again.

If artificial intelligence helps a coder to find information already supported by the record, the resulting may increase reimbursement for the hospital and the hospital should have received all along.

Payers have used data analytics and automated files for years to identify claim denials, payment reductions, and recovery.


Optum describes an advanced analytics for AI for claim review, coding checks, edits, and prevention of overpayments.

Cigna's provider instructions expressly address claims denied because of coding edits.

They also establish that the technology scrutinizing a claim for financial consequences is hardly a new development.

Calling one system data analytics and the other AI does not determine whether its conclusions are correct or incorrect.

Traditional edits and generative AI are a different technology, but both require accountable use.

The suggestion that a second code is inherently suspicious in billing overlooks the structure of the ICD-10 instructions, such as code first and use additional code can require multiple codes and specify their order.

A code also, note, indicates that two codes may be necessary without itself directing the sequence.

For hypertensive chronic kidney disease, for instance, an I-12 category is accompanied by an N18 code identifying the kidney stage.

Similarly, hypertensive heart disease with heart failure requires an additional I-50 category code identifying the heart failure type.

An AI reminder to check for missing companion codes may improve compliance.

These instructions predate AI and conventional coding software can flag them too.

AI can add value by helping locate supporting documentation scattered across a lengthy chart.
The rest of this transcript — segmented and speaker-labeled, so you land on the exact moment something was said
Search every transcript — by keyword, by phrase, or by meaning, across every show Radar indexes
Trends — what is surging across podcasts, measured against its own baseline
Alerts — when a name you follow appears in a newly indexed episode
No account is needed to search Radar.