Prior Authorization Challenges Dental Practices Face (and the Central Billing Office Fix)

In six years as a PM at CareStack, I watched hundreds of billing teams inside the software every day. The same pattern came up at every DSO: eligibility was frustrating, claims were annoying, but prior authorization was where billing directors actually lost sleep. One biller managing a stack of PAs, none of them electronic, none of them on a predictable clock, and a schedule of treatment plans waiting on "approval" that might take a day or three weeks.

This article is for the billing director who has inherited that mess. I'm going to walk through what's actually breaking, why it breaks harder at DSO scale, and the Central Billing Office redesign that gets average PA turnaround from 11+ days to under 3. I'll show you where automation fits, and, just as importantly, where it doesn't.

Prior Authorization vs. Predetermination vs. Eligibility: Get the Definitions Right

Most content conflates these three. Your staff probably does too, which is one reason PAs get submitted when a predetermination would have sufficed (or vice versa). Here's the clean version:

• Eligibility verification answers: Is this patient covered, by which plan, with what benefits, frequencies, and waiting periods? It happens 24–72 hours before every visit. No coverage decision is attached.
• Predetermination answers: If we submit this claim today, how much will the payer estimate covering? It's non-binding. It's an estimate, not a guarantee. Useful for patient financial counseling; dangerous as a scheduling trigger.
• Prior authorization (PA) answers: Will the payer cover this procedure before it's performed? It is a binding coverage decision: when approved, the payer commits to the benefit (subject to continued eligibility at date of service). PA is typically required for crowns, implants, periodontal surgery, orthodontics, sedation, and certain endodontic procedures, though "required" varies by payer and plan.

The real-world consequence: if your biller submits a predetermination when the plan actually required prior authorization, the claim will deny even if the estimate came back favorable. I've watched this exact mistake eat tens of thousands of dollars at a single 12-location DSO. Get the taxonomy right at intake.

The 5 Structural Failure Modes of Dental Prior Authorization

Below are the five things that actually break PA workflows. Fix these and your turnaround drops. Ignore them and no software in the world will save you.

1. Payer Inconsistency: What Requires PA Varies by Payer and Plan

There is no universal PA list in dental. Delta Dental of California may require PA for implants; Delta Dental of Michigan may not. A Cigna DPPO plan may require PA for crowns over a certain code; a Cigna DHMO plan from the same employer may not. Medicaid plans vary state by state and MCO by MCO.

What this means for the biller: PA requirements have to be verified plan-by-plan, at the line-item code level, at the time of treatment planning. "We always PA crowns for Cigna" is a heuristic, not a workflow.

2. Documentation Requirements That Aren't Documented Anywhere

Every payer has a "PA documentation checklist" in theory. In practice, you learn what they want by submitting something incomplete, getting a rejection with a vague reason code, resubmitting, and getting another rejection. For a crown PA, one payer wants narrative + periapical + bitewing.

Another wants narrative + periapical + intraoral photograph. A third wants all of the above plus caries chart notes for the last 12 months.

Your biller finds out by failing. That institutional knowledge lives in her head, not in your PMS, not in a SOP, not in any training doc. When she leaves, it leaves with her.

3. Phone/Fax-Only Submission Channels

Most dental payers still do not accept electronic PA submission through the 278 transaction that medical payers use. Submission is phone, fax, payer portal upload (which is really a manual workflow), or, in the best cases, a dedicated PA form on the payer's provider website.

A biller submitting 20 PAs a day through phone and fax is not doing revenue work. She's doing clerical work that the payer has externalized to you.

4. No Standardized TAT Commitments from Payers

Medical payers have CMS-driven turnaround time rules. Dental payers largely do not. A PA can take 48 hours, or it can take 3 weeks, and the payer's SLA is typically "we'll get to it." Which means your biller has no basis to tell a treatment coordinator when to schedule the patient, and no escalation lever when a PA ages.

The AMA's annual prior authorization survey documents prior authorization as the top administrative burden for physicians in medical settings, dental billing faces the same operational drag without the electronic PA frameworks CMS has required for Medicare Advantage plans.

5. Front Desk Submits PA at Scheduling. Not at Treatment Planning

This is the most common single mistake I see, and it's a timing error. The schedule is built around what the operatory and doctor can produce. PA submission should happen the moment the treatment plan is signed, not when the front desk tries to schedule the appointment two weeks later. Submitting at scheduling costs you 10–14 days of calendar time you already had.

Fix the timing first. Everything downstream improves.

Payer Reality Check: PA Processing Times for Top 10 Carriers

These turnaround times are based on our observations working with DSO billing teams across the U.S. They are not published SLAs, most dental payers do not publish them. Treat them as planning benchmarks, not guarantees. Your experience will vary by state, plan type, and submission channel.

If your DSO's average PA TAT is 11+ days, that is not a payer problem. That is a submission quality problem compounded by follow-up gaps. Clean submissions to the preferred channel move significantly faster than fax dumps.

The Hidden Revenue Leak: Patient Drop-Off Between Treatment Plan and PA Approval

Here's the part nobody puts on the P&L. When PA takes 11+ days:

• Patients reschedule. Their work schedule changes, their co-pay anxiety returns, a competing priority wins.
• Patients forget. Treatment plans signed 2–3 weeks ago lose urgency. The patient doesn't respond to the scheduling callback.
• Patients go elsewhere. If the procedure is painful enough to need doing, and your PA is still pending, a competitor with faster turnaround wins the patient.

In my conversations with billing directors at 10–30 location DSOs, drop-off on PA-required procedures runs 15–25%: meaning one in five to one in four treatment plans signed never gets produced. On a $2M/year crown-and-implant book at a 5-location DSO, that's $300K–$500K of signed revenue walking out the door while a PA sits in a payer queue.

The biller isn't the problem. The workflow is.

Why DSO Scale Makes Prior Authorization Harder

Single-location practices can run PA on tribal knowledge. The biller knows her payers, knows her doctors' preferences, knows which receptionist is good at chasing fax confirmations. It works because everything is in her head.

At a DSO, that model doesn't scale. Here's what I've watched happen at every multi-location group that runs PA location-by-location:

• Distributed teams re-solve the same problem. Ten locations, ten versions of a Delta Dental crown PA package, ten biller interpretations of "what they'll accept."
• Location-level variation in PA quality. Location 3 has a 70% clean-submission rate; location 7 has a 35% clean-submission rate. The CFO finds out six months later.
• Central Billing Office handoffs fail. Location submits PA, CBO handles follow-up, but the chart notes the CBO needs live in the location's PMS record and haven't been synced. Three days lost just getting the documentation to the follow-up team.
• Inconsistent documentation. Dr. A at location 2 writes thorough narratives. Dr. B at location 5 writes two sentences. Same PA, different outcomes.

Scale amplifies bad process. A 3-location DSO can survive ad hoc PA. A 30-location DSO cannot.

The Central Billing Office PA Redesign: What Best-Practice DSOs Do Differently

Here is the redesign I have seen work. It assumes you have a Central Billing Office or are building toward one.

1. Centralized PA Team (Not Location-Level Submission)

Pull PA submission and follow-up into the CBO. Locations own treatment planning and documentation capture. CBO owns submission, follow-up, escalation, and outcome tracking. This single org change is worth 2–4 days of TAT reduction by itself, because it removes the handoff lag.

2. Templated Documentation Packages by Payer and Procedure

Build a matrix: rows are procedures (crown, implant, perio, endo retreat), columns are payers. Each cell is a documentation package: exact attachments, narrative template, chart-note requirements, submission channel. When a new PA is triggered, the CBO biller pulls the cell, assembles the package, submits.

No tribal knowledge. No "let me think about what Cigna wanted last time."

3. Payer-Specific Submission Protocols

Every payer gets a written submission SOP in the CBO. Channel of choice, backup channel, expected TAT, who to escalate to, reference numbers to capture. When a biller leaves, the SOP stays.

4. Proactive Follow-Up Cadence

The worst thing a biller can do is submit a PA and wait. The best DSOs I've worked with run a structured follow-up cadence, status check at day 3, day 7, day 10, with escalation to a payer rep by name at day 14. Aged PA reports run daily, not weekly.

5. Treatment Plan → PA Trigger in the PMS

Treatment plan signature should trigger a PA task in the CBO queue the same day. No dependency on the front desk remembering. Most modern PMSes (CareStack, Open Dental, Dentrix Ascend) can do this with a small workflow config. If yours can't, that's a PMS evaluation conversation.

Where Automation Fits (and Where It Doesn't)

I want to be honest here because the industry oversells this. Prior authorization is harder to automate than eligibility verification: and it will remain so: because PA requires clinical documentation that a machine cannot assemble without human judgment about what's medically relevant to include.

Here is where automation genuinely helps, and where it doesn't:

Where AI/automation helps today:

• Documentation assembly. Pulling the right periapical, bitewing, narrative template, and chart notes into a PA package based on procedure + payer.
• Status checking. Logging into payer portals on a schedule and pulling PA status updates, so your biller doesn't have to. This is the single highest-leverage use case.
• Follow-up triggers. Aged PA reports, auto-escalation flags, payer rep escalation scripts.
• Submission tracking. Capturing reference numbers, submission timestamps, payer confirmation receipts so nothing falls between the cracks.

Where automation does not help (yet):

• Deciding what clinical evidence to include. That's a dentist-and-biller judgment call.
• Writing the clinical narrative. Generic AI narratives get denied. Good narratives reflect the specific patient's condition.
• Handling ambiguous rejections. "Need additional information" requires a human to interpret and respond.

The mistake is thinking automation replaces the workflow. It doesn't. Automation makes a well-designed workflow faster. If your workflow is broken, automating it just breaks it faster.

The 3-Day PA Playbook

If you are a billing director at a DSO and you want to get PA turnaround under 3 days, here is the step-by-step.

Week 1–2: Build the Payer × Procedure Matrix

• List every PA-triggering procedure you run (typically 8–15 codes).
• List every active payer (typically 15–30 at a DSO).
• For each cell, document: is PA required? If yes: what documentation, what channel, what TAT?
• This lives in a shared doc the CBO owns. No one else edits it.

Week 3–4: Centralize Submission

• Move PA submission out of every location and into the CBO.
• Locations still own documentation capture (chart notes, imaging) and treatment plan signature.
• Build the handoff: treatment plan signed → PMS task created → CBO biller picks up within 4 hours.

Week 5–6: Implement Follow-Up Cadence

• Day 3: first status check on every open PA.
• Day 7: second status check; flag any PA with no movement.
• Day 10: escalate to payer rep; document by name.
• Day 14: supervisor review on any PA still pending.

Week 7–8: Layer in Automation for Status Checks

• Identify the 3–5 payers that account for 70%+ of your PA volume.
• Implement automated portal status pulls on a 24-hour cycle for those payers.
• Your biller stops logging in; she gets a dashboard update instead.

Week 9+: Measure and Tune

• Track three metrics by location and by payer: average PA TAT, clean-submission rate, patient drop-off rate between treatment plan and production.
• Aim for: <3 day TAT, >85% clean submission, <8% drop-off.
• Review weekly. Tune the matrix quarterly.

This playbook is not theoretical. I've watched DSOs run it and hit those numbers. What stops most billing directors is not the plan, it's the org change to centralize submission.

A Closing Note to Billing Directors

I built software for billing teams for six years before joining Needletail. The one thing I want to say to a billing director reading this: your PA problem is almost never a staff problem. It is a workflow problem and a process problem, and those are fixable. Your billers are not slow: they are executing a workflow the payers designed to be slow.

When you redesign the workflow, you will be surprised at how quickly the numbers move. 11 days to 3 days is not an aspirational goal. It's a reasonable target at a DSO that commits to the CBO redesign and runs the playbook above. The patients you stop losing to drop-off will pay for the entire initiative within a quarter.

Start with the payer × procedure matrix. Everything else flows from there.

Related Reading

• Dental Claim Denial Prevention
• Coordination of Benefits in Dental Insurance
• Dental Revenue Cycle Management Guide