Prior Authorization in Dermatology

Prior authorization in dermatology is a payer requirement that a dermatologist justify certain treatments before the plan will cover them, and it concentrates on high-cost drugs. The largest driver is biologic and specialty therapy for chronic inflammatory skin disease: agents such as adalimumab, dupilumab, and secukinumab for psoriasis, atopic dermatitis, and hidradenitis suppurativa, billed under HCPCS J-codes and reviewed by payers such as Aetna, Cigna, and Medicare, usually with step-therapy rules attached. Prior authorization is a leading source of administrative work: the American Medical Association reports that practices complete about 39 prior authorization requests per physician per week, roughly 13 hours of staff time. The workflow runs in order: confirm a service needs review, gather the ICD-10 diagnosis and prior-therapy history, submit to the plan or its pharmacy benefit manager, and receive an authorization, a denial with an appeal path, or a request for more information.

What drives prior authorization in dermatology?

The dominant driver is drug cost. Biologics and other specialty agents for psoriasis, atopic dermatitis, and hidradenitis suppurativa carry high annual prices, so plans gate them behind prior authorization and, in most cases, step therapy. Adalimumab, dupilumab, secukinumab, and similar agents are the frequent examples, and the policy typically asks for the diagnosis, a measure of severity such as affected body-surface area or a validated score, and a record of the drugs the patient already tried. Beyond biologics, some plans review advanced oral systemic drugs, extended phototherapy courses, and specific procedures; a subset of plans apply review to Mohs surgery in defined situations. Whether any specific service needs review depends on the plan, the drug, the diagnosis, and the site of care, so the reliable first step for a new biologic order is to pull the plan's drug policy and read its criteria before the office promises the patient a start date.

How does AI handle dermatology biologic prior authorization?

An AI agent carries the repetitive parts of a biologic authorization. It gathers the diagnosis, the severity measure, and the prior-therapy history from the record, assembles the specific evidence the drug policy asks for, and files the request through the payer or pharmacy-benefit-manager portal or the electronic 278 prior authorization transaction where the plan supports it. It can confirm coverage first through the payer's 270/271 eligibility transaction, submit the request through a multi-payer portal such as Availity, hold on the plan's phone queue to check or expedite a pending request, then write the authorization number, approval span, and any quantity limit back into the EHR, whether that is Epic, ModMed, or another dermatology system, so scheduling and the specialty pharmacy can proceed. Because a biologic often routes through a pharmacy benefit manager rather than the medical plan, the agent submits to the right entity and tracks that separate status. It does not decide medical necessity or override the plan's criteria; it gathers and moves information, and escalates a denial or an ambiguous response to staff. Tracking each denial reason is covered in our guide to prior authorization automation.

How does AI manage step therapy for dermatology drugs?

Step therapy is the rule that a patient must try and fail one or more lower-cost drugs before the plan will cover a biologic, and it is where dermatology authorizations most often stall. An AI agent reads the plan's step-therapy criteria for the specific drug, then checks the record for the prior trials the policy requires and their documented outcomes, such as an inadequate response to or intolerance of methotrexate or a topical regimen before a systemic biologic. When a required step is missing or undocumented, it flags that before submission so staff can either document the trial that did occur or start a step-therapy exception request, rather than absorb a denial that was predictable from the criteria. When the record does show the required failures, the agent attaches that history to the submission so the reviewer can confirm the step is satisfied on the first pass. This front-loading matters because a step-therapy denial adds a full appeal cycle to a patient already waiting to start treatment.

Is it HIPAA compliant?

AI prior authorization can be HIPAA compliant, and for this work it must be, because the agent creates and handles protected health information on the practice's behalf. Under HIPAA, a vendor that processes patient records to file authorizations is a business associate, which means the practice needs a signed business associate agreement and the vendor is directly responsible for safeguarding that information, as HHS explains in its guidance on business associates. In practice, compliance means encrypting the clinical data the agent reads and submits, limiting access, logging every action so a person can review what the agent did and why, and signing the BAA before any records are touched. Ask a vendor how it stores portal transcripts and submitted documentation and whether it supports additional standards such as SOC 2. A boundary matters here as much as the paperwork: AI gathers and moves information, but it does not decide medical necessity or override a payer's clinical criteria. Clinical judgment stays with the dermatologist, and a vendor that cannot commit to a BAA should not touch patient records.

How Flexbone helps with dermatology prior authorization

Flexbone runs voice and browser agents, backed by US-based staff through our Nav Central partnership, that carry the repetitive parts of the dermatology prior authorization process. An agent can assemble a biologic request for adalimumab, dupilumab, or secukinumab against the plan's drug policy, check the record for the step-therapy trials the policy requires, file through a pharmacy-benefit-manager portal or the electronic 278 transaction, hold on the plan's phone queue to check or expedite a pending request, and write the authorization number and status back to the record. Each call and submission is logged, so a person can review what the agent did. We do not decide medical necessity or override a payer's criteria; the agents gather and move information, and clinical judgment stays with the practice. Anything they cannot resolve, an ambiguous response or a denial that needs an appeal, escalates to US-based staff.

If you want to map which parts of your dermatology prior authorization workload AI can take off your team, book a call with Flexbone, or see how we approach AI for dermatology more broadly.