Claim narratives that get paid

Narratives are written at the end of a long day, by someone who has already explained the case to themselves, for a reader who has thirty seconds and no context. Almost every improvement available here comes from taking that reader seriously.

What the reviewer is deciding

Not whether the treatment was good. Whether the claim meets the contract: was this procedure necessary, is it covered, does the documentation support the code, and is there a limit that applies. A narrative that does not speak to those four things is being read as filler.

The five things it has to establish

1. The diagnosis, in clinical terms, with the tooth number or quadrant. Not the procedure — the finding that made the procedure necessary.
2. Severity or extent. How much structure was missing, how deep the pocketing was, how many walls remained. Numbers are more persuasive than adjectives.
3. Why the alternative was not appropriate. Reviewers are assessing whether a cheaper covered procedure would have done, and a narrative that pre-empts that question answers the real objection.
4. What supports it. Name the attachments — the radiograph, the chart, the photograph — so the reviewer knows to look rather than assuming there is nothing.
5. Any history that resolves a limit. The date of the previous restoration, the previous prosthesis, or the previous therapy, where a frequency rule is what is actually in dispute.

Why templates stop working

A stored narrative pasted onto every claim of a given type is detectable, and payers detect it. Identical text across many claims is treated as an assertion the practice makes routinely rather than a description of this patient, and it is given the weight that implies.

Worse, it becomes an audit pattern. A practice whose build-up narratives are word-for-word identical across two hundred claims has told a reviewer that the narratives were not written about the teeth.

The workable middle is a prompt rather than a template: a short internal checklist of the five points above, filled in with this patient's findings. It takes a minute longer and it is the difference between a narrative and a signature.

Which claims always need one

• Any by-report or unspecified procedure code. These cannot be adjudicated automatically and are declined without a narrative rather than investigated.
• Core build-ups, where the question is always whether the crown could have been retained without it.
• Scaling and root planing, where charting supports the diagnosis and the narrative explains the extent.
• Anything billed outside a frequency limit, where the history is the argument.
• Two separate restorations on one tooth on one day, which looks like the surface-splitting pattern until something explains it.
• Crowns on teeth with limited visible damage, where the radiograph alone does not carry the case.

Write it at the time, not at the appeal

A narrative written six weeks later is reconstructed from a chart note by someone who was not in the room. It is weaker, it takes longer, and the claim has already been denied once, which changes how the second submission is read.

The practices that do this well capture the narrative at the point of treatment, as part of closing the appointment, while the clinician still remembers which walls were missing. The front office is then writing from a real description rather than inventing one.

Where it goes

The remarks field on the claim has a character limit, and systems truncate silently. Where the narrative is long or where attachments carry the argument, send it as an attachment and put a one-line pointer in the remarks so the reviewer knows it exists. A narrative that was cut off mid-sentence is worse than none, because it reads as carelessness about a claim you are asking someone to treat as an exception.