Biologic Prior Auth in Rheumatology: How AI Cuts the Approval Wait

A rheumatologist prescribes adalimumab for a patient with moderate-to-severe rheumatoid arthritis. The prescription cannot be filled until the payer approves it. The payer requires documentation that the patient has tried and failed two first-line DMARDs — methotrexate and hydroxychloroquine — before they will authorize the biologic. Your staff needs to pull the relevant visit notes, compile the step therapy documentation, submit the prior auth request, wait for a decision, respond to a request for additional information, wait for a second decision, and then call the patient to tell them whether they can get the medication their physician prescribed.

That process takes, on average, 14 days. For a patient with active rheumatoid arthritis, 14 days of waiting for a medication that reduces inflammation and slows joint destruction is not a minor inconvenience.

Rheumatology practices carry the highest prior authorization burden of any specialty in outpatient medicine. Biologic medications — adalimumab, etanercept, ustekinumab, secukinumab, tocilizumab, and a growing biosimilar market — generate prior auth requests that are complex, documentation-heavy, and payer-specific. A rheumatology practice with 3 physicians and 1,000 active patients may be managing 80 to 120 active prior auth requests at any given time. That is not a part-time job for one staff member. It is a full-time job for two.

AI voice agents integrated with athenaOne do not approve the prior auth. That decision belongs to the payer. What AI does is eliminate the administrative overhead that surrounds the determination — the tracking, the status calls, the documentation compilation, the follow-up when a request goes unanswered for a week.

Why biologic prior auth is harder than other specialties

Prior authorization exists across medicine. Rheumatology’s burden is uniquely heavy for three structural reasons.

Step therapy requirements are multi-layer and payer-specific. Most commercial payers require patients to fail one, two, or three first-line treatments before approving a biologic. But the specific requirements vary by payer. United Healthcare’s step therapy ladder for adalimumab is not the same as Cigna’s. Aetna’s documentation requirements for ustekinumab are not the same as Blue Shield’s. A rheumatology practice billing 8 to 12 different commercial payers faces 8 to 12 different step therapy rule sets, none of which are standardized and all of which change periodically without notification.

Biologics require renewal authorizations. Unlike a one-time surgical authorization, biologic approvals typically last 6 to 12 months before requiring renewal. A patient on adalimumab for three years has generated four to six separate prior auth cycles for the same medication. Each renewal requires documentation that the medication is still clinically indicated and that the patient has not developed contraindications. The administrative cycle never ends.

Infusion scheduling adds a second auth layer. Patients on IV biologics — abatacept, tocilizumab, belimumab — require both a medication authorization and a facility or infusion center authorization. Getting both approved in time to schedule the infusion, coordinate with the pharmacy, and ensure the medication is available on the infusion date is a logistics problem as much as an administrative one.

What AI handles in rheumatology prior auth

The components of rheumatology prior auth that AI voice agents and workflow automation address:

Step therapy documentation compilation. AI pulls the relevant prior treatment records from athenaOne — dates of service, medications prescribed, documented failure or adverse event — and compiles them into the format required by the specific payer for the specific medication. The physician reviews and signs. The AI submits.

Real-time status tracking. Rather than a staff member calling each payer every 48 to 72 hours to check on pending authorizations, the AI monitors status through payer portals and electronic auth systems where available, and flags any request that has been pending beyond the payer’s standard turnaround time.

Patient status communication. Patients on biologics want to know when their medication is approved. Currently, many practices have patients calling in to ask for updates — calls that take time, go to voicemail, and generate callbacks. AI handles outbound status calls: when an auth is approved, the AI calls or texts the patient with the update and next steps. When an auth is denied, the AI alerts staff with the denial reason and queues the appeal workflow.

Appeal support routing. When a biologic authorization is denied, the appeal process requires specific documentation: peer-to-peer review requests, clinical justification letters, and in some cases formal appeals with supporting literature. The AI does not write the appeal letter. It identifies that a denial occurred, extracts the denial reason from athenaOne, routes the denial to the appropriate staff member with the denial code and relevant patient context, and tracks whether the appeal has been initiated.

Renewal calendar management. AI tracks authorization expiration dates for all active patients and generates renewal initiation alerts 45 to 60 days before expiration. No biologic patient runs out of medication because a renewal was submitted on day one of expiration.

The infusion coordination layer

Rheumatology practices that run in-office infusion centers have a second prior auth problem on top of the medication authorization. Infusion suite capacity is limited. If a medication auth comes in but the facility auth for the infusion suite is still pending, the appointment cannot be scheduled. If both auths come in but the medication has not been ordered from the specialty pharmacy with enough lead time, the infusion gets pushed.

AI coordinates this sequencing. When a medication authorization is approved in athenaOne, the AI triggers a check on whether the corresponding facility authorization is in place, alerts the infusion coordinator if it is not, and initiates the order workflow to the specialty pharmacy. The goal is to compress the time between “authorization approved” and “patient infused” from 10 to 14 days to 3 to 5 days.

What rheumatology practices report

Practices that have deployed AI-assisted prior auth workflows in rheumatology settings report specific operational improvements:

Staff time on auth administration drops significantly. The high-volume, low-judgment tasks — status checks, patient updates, renewal calendar management — represent 60 to 70% of prior auth staff time. Automating those tasks returns that time to the practice without requiring additional headcount.

Authorization lag time decreases. When AI initiates auth requests the same day the prescription is written (rather than waiting until a staff member processes the queue), and follows up automatically on pending requests rather than waiting for the weekly check-in, the average time from prescription to authorization approval shortens.

Patient satisfaction with the authorization process improves. The single most common patient complaint in rheumatology is not the wait for an appointment or the complexity of the treatment — it is the wait for medication authorization. Proactive AI updates on auth status change the patient’s experience from “I have no idea what is happening” to “I know where my authorization is and what happens next.”

Key takeaways for rheumatology administrators

• Rheumatology carries the highest prior auth burden in outpatient medicine — biologics, step therapy, renewal cycles, and infusion auths compound on each other
• Step therapy requirements are payer-specific and change frequently — AI manages the rule set library so staff does not have to
• Renewal calendar management prevents medication gaps — AI initiates renewals 45 to 60 days before expiration automatically
• Infusion sequencing requires coordination between three auth types — medication, facility, and pharmacy. AI tracks all three
• 60 to 70% of prior auth staff time is automatable — status checks, patient updates, renewal alerts
• Appeal routing is not automated but is structured — AI identifies denials, extracts reason codes, and routes to the right person immediately

The biologic market is growing

The biologics market is expanding rapidly. Biosimilar entry is increasing patient access to biologics that were previously cost-prohibitive, which means more patients in the biologic prior auth pipeline. GLP-1 drugs prescribed off-label for rheumatology-adjacent conditions are adding a second tier of high-scrutiny prior auth to practices that were already at capacity.

Rheumatology practices that are not already running AI-assisted prior auth workflows are managing a problem that is getting larger every year. The administrative infrastructure required to manage biologic authorizations at volume is not proportional to practice size — it scales with the complexity of the payer mix and the number of unique medications in use.

That complexity is not going to decrease. The practices that build automation around it now will absorb the growth. The ones that do not will hire.

Sources:

• American Medical Association, “2023 AMA Prior Authorization Physician Survey”
• American College of Rheumatology, “Prior Authorization Position Statement”
• IQVIA, “Global Oncology Trends and Biosimilar Market Report 2024”

See how Pretty Good AI handles biologic prior authorization for rheumatology practices