An AI phone agent for a dermatology schedule that is booked out for weeks

A medical dermatology practice runs on a schedule that is nearly always full and a phone that never stops testing it. New patients call with referrals in hand and get offered dates weeks out. Established patients call about a mole that has changed, a biopsy they are waiting on, a refill for a medication that needs monitoring, or a prior authorization that is holding up their biologic. Meanwhile every cancellation opens a slot that someone on the waitlist would take within the hour—if anyone had time to work the list. The front desk fields all of it between check-ins, and the calls that slip to voicemail include some that should never wait.

ThunderPhone gives a dermatology practice an AI phone agent that answers immediately, in both directions: inbound coverage for scheduling, intake, refill requests, and status questions, and outbound campaigns for the waitlist calls, recall outreach, and post-procedure check-ins the team never has time to place. The agent works from the practice's own scripts, escalation rules, and approved documents. It is a front-office worker, not a clinician: it never evaluates a lesion, never interprets a symptom, and never delivers a pathology result. Those boundaries are the design.

This page is about medical dermatology. Practices whose phone load is primarily aesthetics retail—injectables, laser packages, memberships—should read the med-spa page instead; the workflows and guardrails differ.

What a dermatology agent can be configured to do

Every workflow below runs on practice-written instructions. The agent follows the script, collects the fields the practice requires, and escalates when a practice-defined trigger is met.

Scheduling against a long waitlist — and backfilling cancellations

When the next routine opening is weeks away, the schedule itself becomes the asset to protect. The agent can book, move, and cancel appointments through a connected calendar or practice-management API, offer to add callers to the waitlist when nothing sooner exists, and record their availability windows. Then it works the list in the other direction: when a cancellation opens a slot, an outbound campaign can call waitlisted patients in the practice's priority order, offer the opening, and book the first taker—so a cancelled 2:40 p.m. does not become idle chair time because the desk was too busy to make six calls.

New-patient intake with referral capture

New dermatology patients often arrive by referral, and the referral details matter for both scheduling and records. The agent can collect the registration fields the practice defines—demographics, contact details, insurance information the practice chooses to take by phone, preferred pharmacy—plus the referral itself: who referred, for what stated concern, and any urgency label the referring office attached. The structured record lands in the intake queue, and referrals the practice has flagged as urgent follow the escalation script below rather than the routine calendar.

Urgent-lesion calls, handled by the practice's script — never by the agent's judgment

Some dermatology calls should not wait for the next routine opening: the caller describing a mole that has changed shape or color, or a severe reaction to a new medication. The practice writes the list of trigger phrases and the exact response—offer a practice-reserved urgent slot, transfer to a nurse, or take a flagged message for same-day clinical callback. When a caller's words match the list, the agent follows that path immediately. What it never does is assess the lesion, ask diagnostic questions the script does not contain, reassure the caller, or decide that something "sounds fine." Urgency handling is a lookup against the practice's written rules, not a judgment.

Biopsy-result callback scheduling — results always come from a clinician

Patients waiting on pathology call often, and those calls are anxious ones. The agent can help with the logistics and only the logistics: confirm the practice's stated result-communication timeline from approved content, take a message for the clinical team, or schedule the callback or follow-up visit at which a clinician will discuss results. The hard rule—stated to the caller in plain language—is that biopsy and pathology results are always delivered by the practice's clinicians. The agent never reads, summarizes, hints at, or confirms the existence of a result, even when the caller pushes ("just tell me if it's bad"). That is a boundary the practice should test explicitly before launch.

Prescription-refill request intake

Dermatology refills frequently involve medications with monitoring requirements, which is exactly why refill calls should produce clean structured requests rather than voicemails. The agent can run a refill line: patient identifiers, medication as stated by the caller, pharmacy, and callback details, delivered to the clinical team through an integration or webhook. It says plainly that the care team reviews every request. It never approves or denies a refill, never discusses dosing, and never speculates about whether required labs or visits are current—those questions route to licensed staff under the practice's protocol.

Pre-procedure instruction calls

Ahead of biopsies, excisions, or other in-office procedures, outbound calls can confirm attendance and read the practice's approved preparation instructions—medications to hold or continue per the practice's written protocol, arrival time, what to expect for aftercare supplies. The agent reads approved content; any question the content does not answer is routed to staff rather than improvised.

Post-procedure check-ins with flagged-answer escalation

After a procedure, the agent can place practice-scripted check-in calls: ask the practice's approved questions about the site, confirm the patient has the aftercare instructions, and record answers. Any response on the practice's escalation list—signs the practice has defined as needing clinical attention—flags the call for human callback. The agent does not interpret the answers; it records and routes them.

Cosmetic-consult scheduling, kept separate from medical

Many medical dermatology practices also offer cosmetic services, and the calls deserve different handling. The agent can recognize a cosmetic inquiry, book it into the cosmetic-consult template with the fields that side of the practice requires, and answer approved questions about consult logistics—without letting cosmetic conversation bleed into medical calls or vice versa. Scheduling rules, providers, and scripts stay separate because the practice configures them separately.

Insurance and prior-authorization status questions

For general questions the practice has approved—accepted plans, how to submit updated insurance, what a prior authorization is and how long the practice's stated process usually runs—the agent answers from the knowledge base. For patient-specific questions—whether a specific prior auth was approved, what a balance is, why a claim denied—it collects the details and routes the call to the billing or prior-auth team rather than guessing. Status questions about biologics and other high-touch medications generate steady call volume; structured routing keeps them from interrupting the desk.

Recall outreach for annual skin checks

Patients the practice tracks for annual full-body skin exams—including those with a history the practice has flagged for surveillance—routinely fall out of recall. Outbound campaigns can work the practice's recall list with approved wording, book the exam in the same call, and handle voicemail through detection plus the campaign's retry policy. The practice remains responsible for ensuring its outbound calling complies with applicable consent rules.

Multilingual patient communication

ThunderPhone supports 47 voice languages. A practice can set a primary language plus additional languages, and the agent switches mid-call when a caller changes to a configured language—so a patient describing a changing mole in Spanish gets the same scripted urgency handling as one describing it in English.

Sample dermatology call flow

Agent: Thank you for calling Ridgeline Dermatology. This is the practice's automated assistant. If you are experiencing a medical emergency, please hang up and dial 911. How can I help you today?

Caller: I need to cancel my appointment Thursday. And I had a biopsy last week—can you tell me if my results are in?

Agent: I can help with the appointment, and I'll explain how results work in a moment. May I have your full name and date of birth?

Caller: Daniel Reyes, June 3, 1969.

Agent: Thank you. I see Thursday at 2:40 p.m. with Dr. Vance. I've cancelled it. Would you like to reschedule now? The next routine opening is in about three weeks, or I can add you to the waitlist and we'll call you if an earlier slot opens.

Caller: Do both, I guess.

Agent: Done—you're booked for the ninth at 10:20 a.m. and on the waitlist for anything earlier. About your biopsy: results are always discussed with you by your care team directly—I don't have access to results and can't share them. I can schedule the results call now, or take a message for the clinical team. Which would you prefer?

Caller: Schedule the call.

Agent: You're set—a member of your care team will call you tomorrow between 3 and 5 p.m. at the number ending in 2214.

Had the caller mentioned a trigger on the practice's urgent list—a rapidly changing lesion, a severe medication reaction—the agent would have stopped and followed the configured path: the urgent-slot offer, nurse transfer, or flagged same-day message, exactly as written.

Platform capabilities behind the dermatology workflows

GDPR and HIPAA compliance posture — and the practice's configuration responsibility

ThunderPhone's public compliance posture is GDPR and HIPAA compliant. That posture does not make any particular practice workflow compliant by itself, and in a medical setting this responsibility deserves the most attention on this page. The practice decides and configures: which patient information the agent may collect and which it must not; whether calls are recorded and what notice callers receive; how long transcripts and recordings are retained; which integrations receive protected health information and under what agreements; what the agent's disclosure script says; and who can access call data. Limit the agent to the minimum information each workflow needs, attach only approved documents, and have your compliance owner review the prompt, recording settings, and data destinations before launch.

Inbound and outbound calling on your numbers

Agents handle inbound calls and place outbound campaigns—waitlist backfill, recall, pre-procedure, and check-in calls all run as campaigns with retry policies. Practices can bring existing numbers through supported direct VoIP connections or manual SIP configuration for any SIP trunk; verified imported numbers support inbound and outbound calling, so patients keep dialing the number they already know.

Warm and cold transfer

Cold transfers route a caller immediately. On eligible production numbers that can place an outbound SIP leg, warm transfers let the agent put the caller on hold, privately brief the nurse or prior-auth specialist, and connect the call only after the recipient accepts—so an urgent-lesion caller is handed to a person, not a queue.

Knowledge grounded in practice documents

Attach the practice's policies, procedure-preparation instructions, accepted-plan lists, and results-communication timeline—text files up to 5 MB, PDF or DOCX up to 50 MB—or import pages by URL. The agent searches these approved sources mid-call instead of improvising, and can be instructed to route any question the sources do not cover.

Integrations that fit the practice's stack

Connected apps, HTTP APIs, and remote MCP servers let the agent act during a call—check the schedule, book against the right visit template, file a refill request, log a waitlist entry. HMAC-signed webhooks deliver call events to practice systems with verifiable integrity. A web widget can put the same agent on the practice's website.

Testing before a patient ever hears it

This vertical demands rehearsal. Run AI-caller simulations of the risky scenarios—a caller mentioning a changing mole mid-reschedule, a patient pressing for biopsy results, a refill caller asking about dosing—with graded scenarios and transcripts. Build the passing cases into regression suites so a prompt edit cannot silently weaken the urgent-lesion script or the results boundary, and use A/B experiments to compare wording on live traffic. Transcripts, recordings where enabled, and grading make production calls reviewable.

What would this cost each month?

Suppose a practice handles 1,200 AI-assisted calls per month, averaging 3 minutes each:

1,200 calls × 3 minutes = 3,600 minutes per month

Spark: 3,600 × $0.02 = $72/month
Bolt: 3,600 × $0.05 = $180/month
Storm: 3,600 × $0.09 = $324/month

That is base engine usage, not a quote. Premium voice or language paths can add up to 3¢ per minute, and phone-number or carrier charges depend on the telephony setup. Hold time bills a flat 2¢ per minute; voicemail and phone-menu minutes bill at the engine rate. Optional add-ons such as verbal acknowledgements, call supervision, and long prompts carry their own published surcharges. The agent builder shows the current all-in rate before calls are placed.

Guardrails a dermatology practice should write into the agent

Each rule belongs in the prompt explicitly, with a simulation that proves it holds:

No clinical judgment, ever. The agent never evaluates a lesion, weighs a symptom, or estimates urgency. Urgent handling is a lookup against the practice's written trigger list, and the response is the practice's written path.
Results come from clinicians only. The agent never reads, summarizes, confirms, or hints at biopsy or pathology results—not even "it looks fine" or "they're in." It schedules the clinician callback and says who delivers results.
No refill authorization. Refill calls produce a structured request for the care team; the agent says so and never predicts the outcome or discusses dosing.
Approved content only. Preparation and aftercare instructions are read from practice-approved documents; anything resembling advice about a skin question routes to clinical staff.
Medical and cosmetic stay separate. Cosmetic inquiries book into cosmetic templates with cosmetic scripts; medical calls never receive cosmetic upsell language.
Minimum necessary information. Collect only the fields each workflow requires; a waitlist call does not need a medical history.
Disclosure and recording notices. The agent identifies itself as an automated assistant and delivers whatever recording or privacy notice the practice's compliance owner requires.
Escalation is always available. Any caller who asks for a person gets a transfer path or a callback message—the agent never stands between a patient and the practice.

Frequently asked questions

Is ThunderPhone HIPAA compliant for a dermatology practice?

ThunderPhone's public posture is GDPR and HIPAA compliant. The practice remains responsible for configuring its own compliant workflow: what patient information the agent collects, recording and notice settings, retention, which integrations receive protected health information, and who can access call data. Review the configuration with your compliance owner before launch.

Can the agent tell a caller whether their mole looks concerning?

No, and it should not be configured to try. The practice writes the trigger list—a changing mole, a severe reaction—and the exact response, such as offering a reserved urgent slot or transferring to a nurse. The agent matches and follows; it never asks diagnostic questions, offers reassurance, or renders an opinion.

Can it give patients their biopsy results?

No. Results are always delivered by the practice's clinicians. The agent can state the practice's approved timeline, take a message, or schedule the callback or visit at which a clinician discusses results—and it tells callers explicitly that it cannot share results, even when pressed.

Can it actually fill cancelled slots from our waitlist?

Yes, when connected to a system that exposes the schedule and the list. The agent records waitlist entries with availability windows during inbound calls, and an outbound campaign can call waitlisted patients in your priority order when a slot opens and book the first taker, with voicemail detection and retry rules governing unanswered calls.

We offer cosmetic services too. Does that conflict with the medical side?

No—configure them separately. Cosmetic consults book into their own templates with their own scripts and providers, and the agent keeps the two conversations apart. If the phone load is primarily aesthetics retail, the med-spa page describes that setup.

Can it answer prior-authorization questions?

It can answer general, practice-approved questions—what a prior authorization is, how the practice's stated process works—from the knowledge base. Patient-specific status questions are collected as structured details and routed to the billing or prior-auth team; the agent does not guess at approval status.

Can one agent serve patients in multiple languages?

Yes. Configure a primary language and additional languages from the 47 supported, and the agent can switch mid-call when a caller changes language, provided the selected voice supports the configured set. Scripted urgency handling applies in every configured language.

How do we verify the agent is safe before patients call it?

Test it like clinical software. Browser mic tests support fast iteration; AI-caller simulations rehearse graded scenarios such as an urgent-lesion mention mid-scheduling or a caller pressing for results; regression suites with pass-rate gates keep prompt changes from weakening the guardrails; and transcripts, recordings where enabled, and grading make live calls auditable after launch.