The Problem With Coding After the Fact
Traditional medical coding happens after the clinical encounter is complete. The provider sees the patient, writes the note, and moves to the next visit. Hours, days, or sometimes weeks later, a coder reviews the note, assigns diagnosis and procedure codes, and the claim gets submitted. The lag between care delivery and coding creates a gap where errors, omissions, and compliance problems take root.
When a coder works from a note written days earlier, they’re interpreting documentation without the benefit of clinical context. If the note is ambiguous, the provider has moved on to hundreds of other patients. Queries to clarify documentation take days to send and longer to resolve. Meanwhile, the claim either waits (creating cash flow delays) or gets submitted with whatever the coder could piece together (creating audit exposure).
This after-the-fact model worked when coding accuracy had lower stakes. Under current enforcement, it doesn’t. The OIG’s BCBS Alabama audit found 91% of sampled diagnoses were unsupported. The Aetna DOJ settlement cost $117.7 million over coding practices that lacked adequate clinical backing. The common thread in both cases: documentation that wasn’t good enough, discovered too late to fix.
What Concurrent Coding Changes
Concurrent coding moves the coding process to run alongside clinical documentation rather than chasing it afterward. As the provider writes the note, or immediately after the encounter closes, AI-assisted tools analyze the documentation and flag coding opportunities and risks in near-real time.
The coder sees the output while the encounter is fresh. If a diagnosis needs additional documentation to support the HCC, the query goes to the provider the same day, not two weeks later. If a code doesn’t match the clinical evidence in the note, it gets caught before submission rather than during an audit. If a condition is clinically present but the provider’s note doesn’t capture it clearly, the system surfaces the gap while the clinical context is still accessible.
This speed matters for accuracy. Studies on documentation quality consistently show that the closer the review happens to the encounter, the more accurate the output. Providers remember clinical details. Coders can ask meaningful questions. Notes get strengthened while the information is still reliable. The further the coding process drifts from the encounter, the more it depends on interpretation rather than evidence.
The Compliance Advantage
CMS and OIG have made it clear that they value encounter-linked documentation above all other forms of diagnosis capture. The OIG’s February 2026 Industry-wide Compliance Program Guidance flagged diagnoses generated from activities disconnected from patient care as high-risk practices. Concurrent coding directly addresses this concern by tying every coding decision to a specific clinical encounter while that encounter is still in progress or recently completed.
The audit trail is stronger because it’s contemporaneous. A code assigned on the same day as the encounter, based on documentation reviewed in real time, with any queries resolved before the note is finalized, presents a fundamentally different risk profile than a code assigned weeks later from a retrospective chart review. Auditors can see that the coding decision was made with full clinical context, not reconstructed from historical records.
For health plans managing RADV exposure, concurrent coding reduces the volume of codes that need retrospective cleanup. Problems get fixed at the source. Documentation gets strengthened before submission. The downstream audit liability shrinks because the upstream process is more rigorous.
Making the Shift
Moving from retrospective to concurrent coding requires changes in technology, workflow, and team structure. Coders need access to notes as they’re finalized rather than in batch queues days later. AI tools need to integrate directly into the EHR to analyze documentation in real time. Provider query workflows need to support same-day turnaround rather than multi-week cycles.
Organizations investing in Concurrent Coding are compressing the gap between care delivery and code assignment to hours instead of weeks. That compression eliminates the interpretation risk, the documentation decay, and the query delays that make traditional coding programs vulnerable to audit findings. The closer coding gets to the encounter, the more defensible every submitted code becomes.
