The Gap Between the Clinical Record and the Final Claim

A home health chart can contain the evidence needed to support a more complete claim, yet some of that evidence never makes it into the final diagnosis list.

Part of the problem is how coding reviews unfold. The primary diagnosis receives close attention. The next few conditions are examined carefully, too. But as a coder works through a lengthy referral packet and a growing list of secondary diagnoses, the depth of review can become less consistent. This isn't necessarily a skills gap. It is simply difficult to sustain the same level of scrutiny across several documents, lengthy records and dozens of coding decisions, especially when turnaround times are tight.

The financial consequences can be significant. Under Medicare's Patient-Driven Groupings Model (PDGM), certain secondary diagnoses can qualify a claim for a comorbidity adjustment. A qualifying condition does not become less relevant to that determination simply because it appears further down the diagnosis list.

For home health agencies, this creates an important question: how do you ensure that the quality of coding review remains consistent from the first diagnosis to the last?

When the Diagnosis List doesn't tell the whole Story

The final diagnosis list is only as reliable as the process used to build it.

A hospital referral may contain diagnoses that are carried forward without sufficient review of the supporting documentation or available specificity. Conversely, a nursing note or therapy assessment may contain relevant clinical evidence that never makes it into the diagnosis list.

In both cases, the challenge is connecting the full clinical record to the final coding decisions—not simply reviewing the diagnoses already listed.

The Payment Rules don't stop at the top of the list

In high-volume coding operations, maintaining consistent scrutiny across every diagnosis is difficult. Complex documentation, guideline cross-checks, interruptions, and fatigue can compound the cognitive demands of the coder’s job. The primary diagnosis naturally receives considerable attention because it plays a central role in determining the PDGM clinical group. Secondary diagnoses deserve systematic review as well, because some can affect reimbursement through the applicable comorbidity adjustment rules.

Medicare's comorbidity adjustment rules don't follow that pattern. Qualifying secondary diagnoses (and, where applicable, qualifying combinations of diagnoses) can affect payment under PDGM even when they appear further down the list. Eligibility depends on CMS's grouping rules, not simply on a diagnosis's position.

The consequences can go beyond a missed comorbidity adjustment. Inadequate specificity, overlooked supporting evidence, or insufficient validation of inherited diagnoses can affect coding accuracy and compliance. The objective isn't to add more diagnoses, but to ensure that every relevant condition is appropriately supported, coded, and evaluated under applicable CMS rules.

Why Coder training alone isn't enough

Refresher training, updated coding guidelines, and targeted quality reviews all have a role in improving coding accuracy. But even well-trained coders must work through lengthy records, reconcile conflicting documentation, and apply complex rules under time and volume pressures.

Traditional automation has limitations, too. A keyword search can find a condition mentioned in a note, but it may not establish whether the condition is current, adequately supported, contradicted elsewhere, or appropriate for coding.

The solution is not to choose between better-trained people and better technology. It is to build a workflow that systematically examines the record, links candidate diagnoses to supporting evidence, applies coding rules, and brings qualified reviewers into decisions that require judgment.

How Serent approaches the problem

Serent.ai combines clinical AI with expert human coders to help home health agencies make diagnosis review more systematic, consistent, and traceable across the entire diagnosis list.

• Reviews the full clinical record: AI-assisted processing helps identify potentially relevant conditions and supporting evidence across referral documents, nursing notes, and therapy documentation.

• Validates diagnoses against the evidence: The workflow helps reviewers assess documentation support, code specificity, and potential inconsistencies that may require clarification.

• Applies coding rules consistently: Candidate diagnoses are evaluated against applicable coding and sequencing requirements, helping agencies capture supported diagnoses without inappropriate additions.

• Keeps experts accountable for the final output: Serent's expert coders review and validate AI-assisted results, applying human judgment when documentation is ambiguous, conflicting, or insufficient.

Consistency Is the real Opportunity

A stronger AI-assisted coding operation with expert human review systematically evaluates the full record, identifies potential omissions, validates inherited diagnoses, and connects coding decisions to supporting evidence. This reduces dependence on where a diagnosis appears in a long list and helps agencies produce more complete, defensible claims.

Serent.ai combines AI-assisted clinical document review with expert coding to help home health agencies close the gap between what the record supports and what the final claim represents.

Reach out to us at hello@serent.ai.