How Markham dental reduced claim denials by 45% overnight

A denied claim rarely announces itself as a problem. It arrives as a line item, gets flagged for rework, and joins a queue that the front desk chips away at between check-ins. The cost is spread so thin across the month that most practices never total it up. At Markham dental, a multi-operatory practice north of Toronto, the team finally did — and what they found changed how the whole office runs.

The problem was never clinical

The instinct, when a claim bounces, is to assume the payer disagreed with the dentistry. The literature says otherwise. In a published review of denied claims, 72.88% of claims are denied due to administrative deficiencies (Miranda GE et al., 2015) — a missing tooth surface, an attachment that never made it out of the imaging software, a narrative that didn't restate what the chart already showed. The clinical judgment was sound; the paperwork around it wasn't.

Preauthorizations follow the same pattern. Research from 2025 found that 68% of preauth denials trace to incomplete documentation (Alencar, 2025). Not wrong treatment. Not an ineligible patient. Incomplete paperwork about correct care.

Dr. Sarah Culling, who practices at Markham, describes the old state of affairs plainly: the documentation was accurate, but it was written for the chart, not for the payer. What an adjudicator needed — the probing depths that justified the scaling, the radiograph reference, the date-stamped narrative — lived in three different places, and assembling it happened after the denial rather than before the submission.

What changed

The practice made two moves, and both were about defaults rather than effort.

First, documentation became audit-ready by default. With Rebrief's agents listening during the appointment, the clinical note is assembled while care is delivered — findings, procedures, and the supporting detail a payer looks for, structured the same way every time. Nobody stays late to reconstruct an appointment from memory; the record that exists at the end of the visit is already the record an auditor would want to see.

Second, claims were scrubbed before submission instead of after rejection. Every outgoing claim is checked against the chart it came from: does the narrative match the charted findings, are the attachments present, is anything the payer will ask for missing? Claims with gaps don't leave the building. The correction happens while the appointment is still fresh, not three weeks later when the remittance comes back.

Neither change asked the team to work harder. Both changes moved the quality gate from after the denial to before the submission.

The morning after

The result gave this piece its headline: a 45% reduction in claim denials at Markham dental, overnight — between the last batch submitted the old way and the first batch submitted the new way. There was no ramp-up period to explain away, because nothing about the practice's dentistry changed. The same clinicians performed the same procedures for the same patient base. The only difference was that the paperwork around that care finally matched its quality.

That is the uncomfortable, hopeful lesson in the number. If denials were mostly clinical, fixing them would be slow. Because they are mostly administrative, fixing them is mostly a matter of never letting an incomplete claim leave the practice — which is exactly the kind of tireless, rule-following work that software should own.

What the front desk does with the time

The denial queue used to function as a shadow job: tracking remittances, pulling charts, writing appeal narratives, re-submitting, re-checking. When the queue shrank by nearly half, that job shrank with it.

Dr. Culling's observation is that the recovered hours didn't vanish into slack — they moved up the value chain. The front desk now spends its afternoons on the work that actually fills the schedule: confirming tomorrow's patients, following up on unscheduled treatment plans, walking patients through their benefits before they sit in the chair rather than after. Collections conversations happen earlier and more gently, because the claim behind them was clean the first time.

There's a morale dimension, too. Rework is demoralizing in a way that new work isn't. An appeal is an argument about the past; a recall call is an investment in next month. Teams feel the difference.

The takeaway

Markham's 45% wasn't a heroic effort. It was a default swap: documentation that is audit-ready the moment the patient stands up, and claims that are verified against the chart before they go out the door. The denial rate fell overnight because the causes of denial — administrative, predictable, preventable — were removed from the pipeline rather than fought in the appeal queue.

The dentistry was never the problem. The paperwork was. And paperwork is a solved problem now.