What psychiatrists and PMHNPs should expect from the next generation of AI documentation
By Allison Sikorsky, DNP, PMHNP-BC
Founder & CEO, PMHScribe
When we talk about AI scribes in psychiatry, we tend to focus on the note. How quickly can it be generated? Does it capture the Mental Status Exam? Can it organize a psychiatric evaluation correctly? Will it save me from charting at night?
Those questions matter. But after years of working in behavioral health, I think they describe only part of the problem.
A psychiatric visit can create an entire trail of work beyond the progress note. A medication may require prior authorization. A patient needs to understand what they’re taking and why. Labs or an EKG may need to be ordered. Psychotherapy provided alongside an E/M service needs to be documented. There may be letters, referrals, or other administrative steps that follow.
Finishing the note doesn’t necessarily mean the work of the visit is finished. That’s where I think the conversation about AI documentation needs to go next.
At its most basic, an AI medical scribe helps turn a clinical encounter into draft documentation for a provider to review. In psychiatry, that can be too narrow a definition.
A single encounter may include medication management, diagnostic reasoning, a Mental Status Exam, risk assessment, psychotherapy, patient education, and follow-up planning. If an AI tool produces a polished progress note but leaves the clinician to manually reconstruct everything surrounding it, it has solved only one piece of the documentation problem.
I learned that firsthand as a PMHNP. The burden wasn’t simply writing notes; it was everything that followed them.
Consider a common medication-management appointment. A patient isn’t responding adequately to treatment. You discuss previous medications, side effects, and why another option may be appropriate. You make a clinical decision together. Then you learn the medication requires prior authorization.
Now someone has to gather the treatment history and clinical rationale for the payer: What has been tried? What happened? Why is this treatment being recommended? Much of that information was already part of the encounter.
Yes. A psychiatry-specific documentation platform can use relevant information captured during an encounter to help create a draft prior authorization letter, including clinical rationale and medication trial history, for the provider to review before submission.
The real opportunity isn’t simply using AI to write another document. It’s using information that’s already there instead of asking clinicians to recreate the same clinical reasoning in another format.
Patient education is another part of the workload that can extend beyond the note.
Psychiatric providers spend meaningful time helping patients understand medications: why something is being prescribed, what to expect, potential side effects, monitoring considerations, and when to reach back out.
After the visit, that conversation may still need to become something useful the patient can take with them.
PMHScribe generates patient-facing psychiatric medication education based on the medication discussed or prescribed during the encounter, which the provider can review before sharing. PMHScribe currently includes medication education in psychiatric evaluations and medication-management visits. The point isn’t to give patients more information for its own sake; they already have access to plenty.
The goal is useful information, clearly communicated.
We often measure documentation technology in minutes saved, and that matters. Clinicians desperately need more time. But there is another burden that’s harder to measure: the mental effort of revisiting the same encounter repeatedly.
A medication conversation becomes part of the progress note. A payer may need the clinical rationale. Psychotherapy needs to be documented. An order may need to be created. A patient may need clarification about the plan.
Each task is reasonable on its own. Together, they can make a patient visit feel like it never quite ends. The next generation of psychiatric AI should reduce how many times clinicians have to reconstruct work they’ve already done.
That’s a more meaningful goal than simply generating a note faster.
Yes. When psychotherapy is provided alongside an E/M service, both the psychotherapy and E/M services must be significant and separately identifiable. PMHScribe supports psychotherapy add-on documentation for 90833, 90836 and 90838 and generates the psychotherapy documentation distinctly from the E/M documentation.
The clinician remains responsible for determining the services performed, selecting the appropriate code, and reviewing the final documentation for accuracy. That’s an important boundary. Technology can support better documentation without making clinical or billing decisions on the provider’s behalf.
Note quality still matters enormously. But in 2026, I’d look beyond the note. Can the platform support psychiatric evaluations and medication-management documentation? Does it understand Mental Status Exams and risk documentation? Can it create distinct psychotherapy add-on documentation when appropriate? Then look at what happens after the note.
Can information from the encounter support patient medication education or prior authorization documentation? Can the platform help with letters, referrals, and orders that psychiatric practices produce every day?
Most importantly, does the clinician remain in control of what is reviewed, edited, and ultimately used? Those questions help distinguish a tool that generates text from one built around the actual workflow of psychiatric care.
I built PMHScribe because I knew what it felt like to finish seeing patients and realize I wasn’t actually finished working. The note was the obvious place to start. But as we’ve built PMHScribe and listened to the clinicians using it, the larger opportunity has become increasingly clear. Psychiatric documentation isn’t a single task.
Today, PMHScribe supports psychiatric evaluations, medication-management documentation, Mental Status Exams, risk documentation, psychotherapy add-ons, psychiatric medication education, prior authorization letters, and practice tools including letters, referrals, and orders.
The goal isn’t to remove clinical judgment from that process; it’s to give clinicians better tools for handling the work their clinical judgment creates.
I don’t think the future of AI documentation will be decided by which platform can generate a SOAP note a few seconds faster; generating text is becoming easier. Understanding what a psychiatric practice actually needs to do with the information from an encounter is harder, and that’s where specialty-specific technology matters.
A good psychiatric AI scribe can help create an accurate first draft.
A better documentation platform understands what comes next.
AI scribes can help psychiatrists and PMHNPs turn clinical encounters into structured draft documentation for provider review. Psychiatry-specific platforms may also support Mental Status Exams, risk documentation, medication management, psychotherapy add-ons, patient medication education, prior authorization documentation and other administrative tools.
Yes. AI-assisted documentation can use relevant information captured during an encounter to help draft prior authorization documentation, including treatment history and clinical rationale. The provider should review and approve the information before submission.
Yes. When psychotherapy is provided alongside an E/M service, psychiatry-specific AI tools can help generate psychotherapy add-on documentation. PMHScribe supports 90833, 90836 and 90838 and generates psychotherapy documentation distinctly from the E/M documentation. The provider remains responsible for coding and final documentation.
Yes. Some psychiatry-specific platforms can generate draft patient medication education based on the encounter. PMHScribe creates medication education based on the medication discussed or prescribed, for provider review before sharing with the patient.
Psychiatric encounters can involve narrative histories, Mental Status Exams, risk assessments, medication management, psychotherapy, and particularly sensitive behavioral-health information. A psychiatry-specific platform is built around those workflows rather than adapting a general medical documentation structure to psychiatry.
Yes. AI-generated documentation should be reviewed by the clinician before it becomes part of the final medical record. PMHScribe allows providers to review and edit generated documentation before copying it into their EHR or practice management system.
Allison Sikorsky, DNP, PMHNP-BC, is the Founder and CEO of PMHScribe and a board-certified Psychiatric Mental Health Nurse Practitioner. Her work spans clinical practice, telepsychiatry, healthcare leadership, medical documentation, and behavioral health technology. She founded PMHScribe after experiencing psychiatric documentation burden firsthand.
See what psychiatric AI documentation can do beyond the progress note. Explore PMHScribe or start a free trial.
