# Nextvisit full content corpus
> AI clinical documentation built for behavioral health.
AI clinical documentation for behavioral health. Notes, codes, and clinical context drafted before the patient leaves the room. Used by psychiatrists, NPs, therapists, and 300+ practices.
Compliance: HIPAA, SOC 2 Type II, BAA. 300+ practices on the platform. Index: https://nextvisit.ai/llms.txt. Sitemap: https://nextvisit.ai/sitemap-index.xml.
Generated at 2026-09-09T05:10:06.511Z.
# Blog
================================================================================
URL: https://nextvisit.ai/blog/what-i-check-before-i-sign-an-ai-note
Title: What I check before I sign an AI note
Description: A draft can be fluent and still be wrong in the places I am responsible for. This is the pass I do before my name goes on it.
Date: 2026-08-19
Author: Faisal Rafiq, MD
Category: Documentation
================================================================================
I like a draft. I do not trust one. The two can sit together if I leave myself two minutes and I look in the same places every time. When I "just skim," I sign other people's mistakes with better punctuation.
This is the pass I do because I have been burned.
## Their words, not a cleaned-up version of their words
I search the note for the sentence that made the visit make sense. The med they stopped. The fight on Sunday. The "I gave my sister the bottle." If that sentence has been smoothed into "patient reports recent interpersonal stress" I put the actual words back, or I do not sign.
A tool that keeps a verbatim line is doing me a favor. A tool that paraphrases the only line that mattered is doing me a harm I will not notice if I am late.
## Meds: what I said I would do, what the note says I did
I look at the medication list and the plan as if I have never seen this patient. Dose, change, why. If I told them we would hold the increase until the TSH is back, and the plan says "increase as discussed," I have created a different visit.
I also look for extras. Drafts sometimes keep a med I tapered off in June, or drop a PRN I mentioned once. I have signed both errors. The patient then lives in the error.
## The exam I performed, not the exam the template wanted
If I did a short exam, the note stays short. I will not accept a full, balanced MSE I did not do, even if it is flattering. I already wrote about copy-forward. Same disease, new host.
If something was off, it has to be in the exam and in the assessment. An agitated person with an exam that says "calm, cooperative" is a note I cannot defend, and I should not need a reviewer to tell me that.
## Necessity in this visit, not in the diagnosis list
I read the first problem, the assessment, and the plan out loud in my head. If I cannot hear why they needed me today, I write the sentence. "Still not sleeping, still missing work, last increase was 12 days ago, giving it time" is a reason. "MDD, continue meds" is a stamp.
Scoring a PHQ-9 does not finish this. The score has to agree with the story or I have to say why it does not.
## Codes last, and only against the note I now believe
I used to pick the code and then hope the draft would dress for it. That is backwards and it is how you get a 99215 that reads like a refill. I finish the note. Then I ask what I actually did. Time, complexity, psychotherapy content that is really there.
If the code I wanted needs a sentence I do not have, I either add the sentence because it is true, or I pick a smaller code. I do not invent the sentence.
## What I skip
I do not line-edit every transition. The note does not need to sound like a journal.
I do need my name to mean I was there, I looked, the dangerous parts are right, and I would still make this plan tomorrow morning.
Two minutes. Same places. Then sign. If I cannot spare the two minutes, I do not have time to see the next patient either. I just like pretending I do.
================================================================================
URL: https://nextvisit.ai/blog/stop-copy-forwarding-the-mental-status-exam
Title: Stop copy-forwarding the mental status exam
Description: A pasted MSE is how a stable paragraph survives a week that was not stable. Here is what I still write myself, and what I let a draft start.
Date: 2026-07-22
Author: Faisal Rafiq, MD
Category: Documentation
================================================================================
I can date a bad habit in my own charts. Around year four of practice I got fast, and "fast" meant the mental status exam from last month survived with a new timestamp. I told myself I would have noticed if something had changed.
I missed a lithium tremor for two visits that way. The patient had it. I had a paragraph that said he did not. The paragraph was prettier than the patient.
## Copy-forward is a different error from a short exam
A short MSE on a known, stable person is fine. "Alert, cooperative, speech normal rate, mood 'tired,' affect congruent, no psychosis elicited, cognition grossly intact." That is a human who looked and then stopped writing.
Copy-forward is a paragraph with grooming, eye contact, thought process, proverb interpretation, and a fund of knowledge, identical to April, on a day they came in unwashed and talking over you. I notice the identical ones when I review a chart. A lawyer will notice them faster.
The tell in a stack of my own old notes was the same clause on thought process, word for word, including a comma I always put in the same place. Nobody's thought process is that loyal.
## What I still insist on seeing this visit
Appearance if it would photograph differently than last time. If it would not, I do not write a fashion paragraph.
Speech and motor, because those change on meds and on mania and on Parkinsonism, and they change before the patient has a speech about it.
Mood in their words, in quotes, even if the quote is "fine." My paraphrase is not their mood.
Affect as I saw it, especially when it does not match the quote. That mismatch is often the exam.
Thought content I actually asked about. If I did not ask about SI, I do not get to write "denies SI" because last month's note did. I ask, or I write that I did not.
What I no longer manufacture: a full cognitive exam I did not do, a proverb I did not ask, "insight and judgment fair" as a reflex. Fair to whom.
## Drafts make this better and worse
A listener in the room will often catch the sentence I would have used as the exam. "I haven't slept in four days" belongs in the note the way they said it. That is useful. I will keep that.
The same draft will also offer a complete, symmetrical MSE because templates like symmetry. If I am tired, I will accept it. That is how I get last year's problem back, except now it is fluent.
My rule now is ugly and it works. I delete the MSE block and write six lines from memory of the last five minutes of the visit. If I cannot, I was not examining, I was chatting. Then I go look at the draft and steal only the phrases that match what I just wrote.
If a tool surfaces the exact line they said, I keep the line. I still do not keep the rest of the paragraph as a gift.
## The new style of bad chart
We have had ambient drafts long enough that I see this more: every section filled, every visit, including the exam, with the same temperature. It looks responsible. It reads like nobody was in the room.
I would rather sign a short exam I performed than a long exam I inherited. I do not open the chart to look competent. I open it to remember what this person was like on this day. Copy-forward is me refusing to look.
================================================================================
URL: https://nextvisit.ai/blog/when-not-to-use-an-ambient-scribe
Title: When I would turn the ambient scribe off
Description: A listener in the room is the wrong tool for some visits. These are the ones where I tell people to type, or to wait.
Date: 2026-06-24
Author: Ryan Yannelli
Category: Operations
================================================================================
I spend most of my year telling practices to stop typing after 7pm. This is the other half. There are visits where a listener costs you more than it saves, and the clinicians who look reckless are the ones who never turn it off.
Default on is fine. Default on with no off switch is how you get a complaint, or a note that looks complete because the machine was confident.
## The patient said no, or hesitated
This is not a debate. Off. If you have to sell them on the microphone, you have already changed the session. Type the note. You knew how to do that last year.
Hesitation counts. People-pleasing patients will say yes while they watch the phone. If the yes is thin, take it as a no for today and ask again another time, or never.
## Someone else is in the room who did not agree
Partner on speaker. Parent in clinic for a 24-year-old. Translator who is a cousin. Group. The consent you got from the identified patient does not cover the other voices. Either get an explicit yes from everyone who can be heard, or the scribe stays down.
I would rather explain a typed note than explain why an ex-spouse's rant is in a medical record.
## The visit is not a clinical conversation
Prior auth on hold. You are reading a lab portal and leaving a voicemail. Collateral-only call with a case worker while the patient is at work. Refill that took four minutes and a checkbox.
An ambient draft of that will either be empty or inventive in the small, irritating ways: a "session" that was a logistics call, an MSE built from hold music and your side of a phone tree. Type the four lines. Do not launder a callback into an encounter.
## You need the room to feel like there is no audience
Some trauma work. A first disclosure they have not said out loud before. The clinical task is to make the room smaller. A visible listening device makes it bigger. Even when the patient consented last month.
You can still document. You just do it the old way, and you write less, on purpose. A short note that is true is better than a fluent note of a session the patient will regret if they ever request records.
If you use Nextvisit, this is a session-level decision. Turn it off for this one. Do not make the patient petition you.
## You cannot review before you sign
If you are stacking visits so tight that drafts will sit unsigned until Sunday, you do not have an ambient workflow. You have a backlog with better grammar. Off, or fewer patients. I mean that.
A draft you never read is not documentation you did. It is a plausible story with your name on it. That was already a bad idea in February. It is a worse one now that the drafts are smoother.
The messy med-management hour is still where a listener earns the seat. The med they stopped two weeks ago. The sentence you would have lost if you had looked down. Those are the visits I want captured.
Use it as a default, not a personality. The clinicians I trust with it are a little trigger-happy on the off switch. That is the opposite of a problem.
================================================================================
URL: https://nextvisit.ai/blog/the-intake-note-is-where-the-chart-goes-wrong
Title: The intake note is where the chart goes wrong
Description: Follow-up notes inherit the first visit. If the intake is thin, later denials and confused covering clinicians start there.
Date: 2026-05-13
Author: Ryan Yannelli
Category: Documentation
================================================================================
I can usually tell, in five minutes, whether a practice has an intake problem. I open one established patient at random and look at visit one. If that note is a sketch, the rest of the chart is a sketch with better dates.
Follow-ups get the attention because they are the daily grind. The intake is the document everyone else will trust. Covering clinicians. Auditors. The therapist you add in month four. The NP who sees them when you are out. They will not re-take the history. They will copy what you wrote.
## What a thin intake actually costs you
A missing prior med trial becomes a second trial of the same SSRI next spring. Skip a family history of bipolar and you start an antidepressant you would have held. Skip the substance line and you write a benzo you would have thought twice about. None of that shows up as "intake quality" on a dashboard. It shows up as a messy year.
Payers do this too. Medical necessity for visit 14 is often justified by the story you told in visit 1: duration, function, failed outpatient, why this level of care. If visit 1 says "anxiety, start sertraline," visit 14 has nothing to stand on except habit.
I have watched groups spend weeks arguing about AI note quality on follow-ups when the real problem is that the first visit never established the patient.
## The sections people rush
History of present illness that starts at this week. You need onset, course, what they already tried, and what "better" would mean. If they have been depressed since the divorce in 2019 and you write "feeling down lately," you have invented a new illness.
Prior meds without doses or why they stopped. "Tried Lexapro, didn't like it" is how you repeat Lexapro. "Escitalopram 10 mg, two weeks, agitation, stopped" is a fact.
Allergies and adverse effects, separately. People say "allergy" when they mean nausea. The next prescriber will treat that word as gospel.
Social and substance in one shrug. Housing, weapons, cannabis daily, alcohol "socially" that is a six-pack. This is the stuff that changes the plan and never appears in a 12-minute intake because the clock won.
A risk section that is only "denies SI/HI." If you did not ask about plan, means, prior attempts, and what they would do this weekend if it got worse, you did not assess risk. You cleared a checkbox.
## You cannot finish an intake in the same shape as a follow-up
This is where tools get blamed. An ambient draft will follow the visit you ran. If the visit was a long med-management appointment pretending to be an evaluation, the note will look like a follow-up with a new patient name.
Block the time. Tell the front desk an intake is an intake. If you only have 25 minutes, you do not have an intake. You have a first prescription and a debt you will pay later.
I would rather see a planned second visit to finish history than a signed chart that pretends you finished. Write "developmental history incomplete; scheduled" and mean it. Do not paste a normal childhood you did not take.
## What I look for when I spot-check
Open the intake. Can I tell why they came, what they have already tried, what is dangerous, and why you started what you started? If I can, the later notes have a spine. If I cannot, no scribe and no template will save the follow-ups. They will only produce cleaner versions of a missing story.
Nextvisit can draft from the visit and drop the result into whatever intake template you already use. That is useful when the visit itself was an evaluation. It will not grow history you never asked. The first visit is still the one you have to run like it matters, because every other note in that chart is going to believe you.
================================================================================
URL: https://nextvisit.ai/blog/what-golden-thread-means-in-behavioral-health-notes
Title: What payers mean by a golden thread in behavioral health notes
Description: Golden thread is not a slogan. It is whether a reviewer can follow this visit from the problem, through what you did, to why the next visit is still necessary.
Date: 2026-03-25
Author: Ryan Yannelli
Category: Operations
================================================================================
If you have been through a behavioral health audit, you have heard "golden thread." People nod. Then they go back to writing notes that list a PHQ-9, a diagnosis, and "continue current treatment."
The thread is simple. A stranger who only has the chart should be able to answer: what is going on, what you did about it today, and why this still needs a licensed clinician. If any of those is missing, the note may be true and still fail.
I am not a psychiatrist. I sit with the people who get the recoupment letters. This is the pattern in those letters.
## The chain, in order
Start with the problem as it showed up today. Not the lifetime diagnosis dump. "Sleep still broken, missing two days of work, sertraline 50 mg for six weeks with partial response" is a problem. "MDD, recurrent, moderate" is a label. You need both. The label without today is how you get "not medically necessary."
Then the assessment has to do work. Why this, still, now. What you are ruling out. What would change your mind. A sentence of actual thinking is worth more than a copied criteria list.
Then the intervention has to match. If you billed psychotherapy time, something psychotherapeutic happened and you can say what. If you changed a dose, the note says why this dose, not just the new number. If you did nothing, "nothing" needs a reason: the last change was eight days ago, side effects are settling, they have a therapy intake Friday.
Then the plan has to point at the problem you opened with. Follow up in four weeks is not a plan if you did not say what you will look for. "Recheck sleep and work function; if no movement, discuss augmentation" is a plan.
That is the thread. Problem, thinking, what you did, what happens next. A reviewer pulls one visit and walks it. They do not owe you credit for the excellent work in your head.
## Where notes usually break
The assessment is last month's paragraph with a new date. The plan is a standing sentence. The PHQ-9 dropped four points and nobody mentioned it. You billed 90833 and the body of the note is meds and "supportive conversation."
Therapists break it a different way. Beautiful process notes, thin medical necessity. "Client processed anger toward mother" can be true and still not tell a payer why this is treatment rather than a good talk. The treatment-plan goal and how they are functioning have to show up in the same note as the intervention.
Psychiatrists break it by writing a perfect med list and a canned MSE, then leaving the "why this visit" implied. Implied does not survive a reviewer who is paid to doubt you.
## Scores do not replace the thread
I like scales. Practices that never record a PHQ-9 or GAD-7 look sloppy. Practices that paste the score and skip the sentence about what it means also look sloppy. "PHQ-9 is 11, down from 16, sleep item still a 3, that matches what she described" is a thread. "PHQ-9: 11" is a number.
Same for risk. "Denies SI" on every note, including the one where they talked about giving the pills to a friend "for safekeeping," is how you lose credibility for the whole chart.
## What this has to do with AI drafts
A draft can lay the sections out and still miss the thread. Fluency is not a through-line. If the visit never connected the problem to the plan, the draft will be a tidy version of that failure.
Read the note the way an auditor will: start at the chief complaint, jump to the assessment, jump to the intervention, jump to the plan. If you have to invent the connection, edit until you do not.
We build Nextvisit to draft from the visit and to use the template you actually bill against. That helps with completeness. It does not bless a note that never makes an argument. You still have to make the argument. The point of a faster draft is that you have time left to do that before the next knock on the door.
================================================================================
URL: https://nextvisit.ai/blog/ambient-scribe-vs-dictation-vs-typing
Title: Ambient scribe, dictation, or type it later: what actually changes
Description: Ambient listening, dictation, and typing after clinic are different jobs. Here is how each fails in a psychiatry or therapy day, and when I would still pick it.
Date: 2026-02-18
Author: Ryan Yannelli
Category: Operations
================================================================================
People shop "AI scribes" as if the category is one product. It is not. Most of the disappointment I hear is a mismatch: they bought a listener and still work like a typist, or they bought dictation and expected it to remember last March.
This is the comparison I walk through when a practice is trying to decide what to change first.
## Typing it after the visit
You already know this one. The patient leaves. You reconstruct. If you are honest, the note is a mix of what happened, what you always write, and what you can still remember at 6:40.
It is private. No microphone, no consent speech, no vendor. It also punishes you for being a good listener. The better the session, the worse the note, because you were not taking minutes.
I still type when the visit is short, the patient declined a listener, or I need one paragraph and I already know what it is. I do not type a 50-minute intake at the end of a clinic day and pretend that is quality.
## Dictation (including the "AI" kind that waits for you to talk to it)
Dragon-style dictation, and the newer tools that clean up your speech, are faster typing. You still do a second shift. You stare at the chart and narrate the visit to a machine.
That is a real upgrade if your bottleneck is fingers. It is not an upgrade if your bottleneck is that you cannot hold the whole hour in your head. Dictation will faithfully write down a confident, incomplete story.
It also keeps you in the note. Some psychiatrists like that. They think in the assessment while they talk. Therapists who already hate the EHR usually do not want another ten minutes of talking to it.
Use dictation when you already know the structure and you are filling a template you trust. Do not use it as a memory prosthetic. It cannot recall a sentence you did not say into the mic.
## Ambient listening
A listener in the room (or on the telehealth call) drafts from what was actually said. The source material is the visit, not your recollection of the visit.
That helps on long, messy encounters: the patient who circles back to the sleep problem at minute 38, the med list correction, the exact wording of a safety plan. It does less for you if the visit was mostly silence, or mostly paperwork, or you spent the hour in a collateral call the patient was not on.
You pay for that with process. Consent. A BAA. A review step, because a fluent draft can still be wrong. If you skip the review, you have a well-written note you did not write. That is worse than a sloppy note you did write.
Ambient tools also inherit the visit you ran. If you never asked about side effects, the draft will not invent a side-effect review. Some people call that a bug. I call that the tool refusing to lie for you.
## How I would choose, in a real week
Most clinicians I talk to should not pick one mode for every visit. They should pick a default.
If your evenings are eaten by intakes and med-management follow-ups, ambient first. Review before the next patient, or before you stand up. If you let those drafts sit overnight, you will edit them like they are someone else's chart, because they will be.
If your panel is mostly 20-minute med checks you could write in four minutes, dictation or a tight template is fine. Buying a listener so you can narrate a refill is theater.
If a patient does not want a microphone, type. Do not sulk. The visit still counts.
Nextvisit is in the ambient camp: it listens, it drafts a structured note, you edit and sign, and the note can go into the EHR you already use. That is the job. It does not replace dictation for people who think by talking to the chart, and it does not make a bad visit look complete.
Decide when the note gets written, and what it is written from. Memory, your spoken summary, or the visit itself. Those are different records.
================================================================================
URL: https://nextvisit.ai/blog/what-belongs-in-a-telehealth-psychiatry-note
Title: What belongs in a telehealth psychiatry note that an office note can skip
Description: Payers and boards still treat video visits as a different setting. These are the fields I see missing when a telehealth chart gets pulled.
Date: 2026-01-14
Author: Ryan Yannelli
Category: Documentation
================================================================================
I keep seeing telehealth notes that would be fine if the patient had been in the office, and thin if anyone asks where the visit actually happened. The clinical work is the same. The record is not.
These are the extras I hear about from billers and medical directors when a video visit comes back.
## Location is not optional
Write where the patient was, and where you were. City and state is the usual minimum. "Telehealth" in the header is not a location. If they were in a parked car outside a pharmacy in a different state from last month, that belongs in the note, because your license might care.
If you do not know where they are, ask before you start. I have watched people realize, twenty minutes in, that the patient took the call from a state they are not licensed in. That is a much worse problem than an awkward opening question.
## Consent for this visit, this modality
A one-time telehealth consent in the chart from 2023 is better than nothing. It is not the same as noting that this visit happened by video and the patient agreed to proceed that way today. If the connection dropped and you finished by phone, write that. The modality changed. The note should change with it.
Some states still want you to document that you told the patient the limits of a remote exam. That can be one sentence. Skipping it is how a clean visit becomes a documentation finding.
## What you could not examine
This is the part office notes get for free. You smelled alcohol. You saw them walk down the hall. On video you have a rectangle and whatever they point the camera at.
Say what you saw and what you could not see. "Patient appeared from the shoulders up, seated, in a quiet room. Gait and tremor not assessed on video." That is a better MSE than a full canned paragraph copied from last month's in-person visit.
If a kid kept walking through, or they took the call from a worksite bathroom, write it. Those details explain why the exam is thinner. They also explain why you scheduled them back in person.
## Time, and what the time was spent doing
Telehealth does not relax time rules. If you bill on time, the note still needs start and stop, or a duration that a human would believe. If you billed an add-on psychotherapy code, the note still needs psychotherapy content, not just "supportive discussion." Video does not fill that in for you.
I would rather see a slightly short, honest note than a 99215 with a video visit that reads like a refill callback.
## Safety when you cannot keep them in the room
Ask where they are, whether they are alone, and what you would do if this went badly. You do not need a paragraph. "Patient at home in Austin, sister in the next room, crisis plan reviewed" is a real sentence. "Safety discussed" is not.
If you are worried, say what you actually did. Stayed on the line. Called a contact. Vague safety language on a remote visit is worse than on an office visit, because you already had less control.
## The boring header fields people skip
POS 10 versus 02, depending on the patient's site. The platform used, if your group cares (Zoom for Government is not FaceTime). Any other person on the call, and why. Interpreter, if you used one.
None of that is clinical poetry. All of it is what a reviewer looks for when they are trying to decide whether the visit happened the way you billed it.
You can keep a short telehealth block at the top of the template and stop thinking about it. The practices that do not do that spend January fixing notes from November. I would rather they spend January seeing patients.
================================================================================
URL: https://nextvisit.ai/blog/do-you-need-consent-for-an-ai-scribe
Title: Do you need patient consent to use an AI scribe?
Description: If a tool is listening to the visit, treat it like a recording. Here is the consent conversation I tell practices to have before they turn one on.
Date: 2025-12-10
Author: Ryan Yannelli
Category: Compliance
================================================================================
Someone asks this on almost every demo, usually after they have already watched a note come together. Then it lands: a microphone was in the room.
I am not your lawyer, and your state might not match the next county over. What I can tell you is how practices that do this carefully talk about it, and where people get sloppy.
## Treat it like a recording, because it often is
An ambient scribe listens to the visit and turns speech into a draft note. In a two-party consent state, that is close enough to "recording a conversation" that you should not wing it. One-party states are looser on the statute and not looser on professional judgment. Therapy and psychiatry add a second layer: even when the law allows you to proceed, the patient still gets to decide whether they want a device in the room.
If your counsel already wrote a recording policy for trainees, telehealth, or quality review, start there. Do not invent a separate AI policy that contradicts it.
## Consent is a conversation, not a checkbox buried in intake
The practices that get fewer surprises say it out loud, in ordinary language, at the start of the first visit that uses the tool:
We use a documentation assistant that listens so I can stay with you instead of typing. It drafts the note. I read and edit that draft before anything is signed. You can ask me to turn it off for this visit, or for part of it.
That is the whole speech. If they hesitate, turn it off. You can type. You will not die.
Write down that you discussed it. "Patient informed that an ambient documentation tool would be used; agreed" is enough for most charts I have seen. If they decline, write that down too, and do not argue them into it.
A portal blurb and a signed general consent help. They do not replace saying it in the room. People skim portals. They remember what you told them while they were sitting there.
## What patients usually want to know
First they want to know who hears this. Name the vendor. Say whether a human on that side can listen (most clinical scribes used for notes do not send your session out for a person to transcribe). If you have a BAA with that vendor, say you do. If you do not have a BAA, stop and get one before you turn anything on. That part is not optional.
Then they want to know whether the session trains some model. Ask your vendor in writing. Put the answer in your own words for the patient. If you cannot get a straight answer, that is your answer.
Then they want to know if they can say no. Yes. Mean it. The visit still happens.
I would not lead with encryption speeches. Patients hear "we take privacy seriously" from everyone. They want to know who is listening and whether they have a choice.
## Special cases people forget
Minors. The parent who consented to treatment is not automatically the person who can consent to a recording, depending on the state and the child's age. Ask counsel before you assume.
A second person on speaker, or a family member in the room. They did not sign your intake. Tell them the tool is on, or turn it off.
A patient who is manic, floridly psychotic, or otherwise not in a position to take this in. Use judgment. You can document later. You cannot un-record a session.
Group therapy. One yes does not cover the circle. If anyone in the group does not want it, the scribe stays off.
## What I would not do
Do not hide the tool and hope nobody asks. A line in 8-point font on page six of the new-patient packet is not informed consent. A vendor marketing page is not your policy.
Nextvisit is an ambient documentation tool. It drafts. A clinician still reviews and signs. You are still asking permission for a listener in the room. You are not asking them to let a stranger write the finished chart.
If you want a starting point for counsel, give them this: we will disclose that a documentation assistant listens, we will document the patient's choice, we will honor a no, and we will keep a BAA on file. Then let them write the actual policy. That is their job. Showing up in the room and saying it plainly is yours.
================================================================================
URL: https://nextvisit.ai/blog/ai-bridges-experience-gap-young-mental-health-clinicians
Title: What new clinicians used to need years of volume to notice
Description: New psychiatrists and therapists get pattern flags, documentation checks, and longitudinal context that used to take years of volume to notice.
Date: 2026-05-08
Author: Faisal Rafiq, MD
Category: Operations
================================================================================
The first years out of training are a lot of panel, a hostile EHR, and payer rules that punish a thin note. AI does not give a new psychiatrist or therapist ten years of judgment. It puts pattern flags, documentation checks, and the patient's own history on the screen while they still have to decide.
## What experience usually is
Textbooks do not teach an incomplete risk exam, or the feel of a diagnosis you are not sure of. That comes from volume. High caseloads and strict notes leave less room to get that volume slowly.
"AI accelerates the learning curve by surfacing patterns across patient histories that might take years for clinicians to recognize independently."
A system that reads the encounter can flag a missing safety line, a diagnosis that does not match the story, a drug interaction, or a lab that is due. Those are the misses early-career clinicians make when they are late and still trying to look sure.
The same read across time can surface early mania, a med that never worked, a slide toward decompensation, or trauma-related behavior that one visit does not explain.
Notes are where new clinicians get hurt. A check that the draft meets the payer, supports the code, includes risk, matches criteria, states a med rationale, names functional impairment, and has a measurable goal is how you avoid the audit you cannot afford in year two.
Form fill, coding checks, and chart review are also the work that steals the hour you needed for the interview. Experienced people already protect that hour. New people often do not know they are allowed to.
When one supervisor covers several trainees, the same flags keep the standard even: which cases need a look, which notes are thin. A rural clinic without a deep bench gets the same prompts as an academic shop. Geography should not decide the quality of the first two years.
Access to a differential, a treatment algorithm, safety-screening prompts, and a monitoring checklist does not replace supervision. It means the first guess is less lonely. Confidence here is not swagger. It is having the file and the protocol in front of you when the attending is in another room. The attending still signs the thinking.
================================================================================
URL: https://nextvisit.ai/blog/ai-clinical-memory-timelines-event-tracking-mental-health
Title: The covering clinician should see what you already know
Description: AI patient timelines keep med trials, life events, and risk history available for the covering clinician, not only the person who wrote the first note.
Date: 2026-05-08
Author: Faisal Rafiq, MD
Category: Product
================================================================================
The hard part of mental health care is often memory. A full-time clinician sees thousands of encounters a year: trials, life events, symptom patterns, what actually helped. Nobody holds all of that in working memory. The cost is repeated questions, late interventions, and a covering clinician who is guessing.
AI is useful here when it builds a patient timeline, keeps the events that change the plan, and lets the practice see the same history the primary clinician thinks they remember.
## Life events that belong on the next visit
AI can pull significant events out of the notes and surface them when they matter again. A September visit about coping can show that last September was the anniversary of a trauma. Elevated mood in early summer can sit next to three prior summers that included hypomania.
That is empathy you can schedule: you did not forget what they already told you. It is also diagnosis. Current symptoms get a history instead of a blank slate.
You spend the visit with the person, not rereading last March in the middle of the hour.
## Cross-coverage
A covering clinician walks in cold: no feel for baseline, past responses, or risk. That produces defensive medicine, the same history taken twice, and the warning sign only the usual prescriber would have caught.
A timeline that opens in a minute should show medication history with responses and side effects, recent risk, how they decompensate, current stability, pending labs or open tasks, relevant social context, and symptom trend. The patient should not have to re-explain the last two years because you are on call. The practice also reduces the liability of a coverage night where the dangerous fact lived in someone else's head.
## Instability before it is a crisis
The useful read is change across visits: rising irritability, sleep disruption, slipping adherence, early activation, a downshift in mood, more PTSD symptoms, or a rising chance of substance relapse.
Sleep plus mild irritability can be a dose conversation. Waiting for mania can be a hospitalization. Experienced clinicians do this from familiarity. A timeline does it for the person who has never met the patient.
## Eligibility for the next step
The same memory can watch for criteria: treatment-resistant depression that may meet esketamine (Spravato) rules, depression still present after adequate trials that may warrant TMS, recurrent severe depression that may need an ECT conversation, response patterns that suggest pharmacogenetic testing, or a protocol such as TMS for bipolar depression.
Two adequate antidepressant-class trials without a response is the moment to raise Spravato or TMS, not the moment after you have tried every remaining oral option from habit.
## The panel, not only the patient
Practice-level views find people overdue for metabolic labs, at higher relapse risk, possibly eligible for advanced treatment, or lost to follow-up. They also show which meds work in this clinic, which side effects stall care, and how long stabilization usually takes in each diagnosis. That is how you staff, how you write a protocol, and how you talk about quality in a value-based contract, instead of arguing from the last case you remember.
## Transitions
Patients move, switch clinicians, leave, and come back years later. A timeline that survives those breaks means the new clinician starts from the record, not from a first-visit fiction. Whether they return after five years or switch inside the group, the treatment history should still be there. Continuity stops depending on one person's memory.
================================================================================
URL: https://nextvisit.ai/blog/ai-clinical-supervision-healthcare-training-hipaa-compliance
Title: Supervision between sessions, without leaking PHI
Description: Trainees need feedback between supervision sessions. A HIPAA-safe review of drafts beats Reddit threads and group texts.
Date: 2026-09-03
Author: Faisal Rafiq, MD
Category: Compliance
================================================================================
Training psychiatrists, therapists, nurse practitioners, and physician assistants still depends on a supervisor who also has a full panel. When that feedback is slow, trainees invent a channel: texts, social groups, or other places that were never built for PHI. A HIPAA-safe tool can sit in that gap without replacing the attending.
## The gap in the training day
New clinicians get less face time with attendings than the generation that trained them. A hard encounter is fuzzy by the time supervision happens. Documentation rules move. Risk language is what keeps people up. In that vacuum people search Reddit, post a case in an informal group, or ask on a non-HIPAA thread.
Healthcare-grade systems keep the question inside encryption. Clinicians can work from de-identified summaries. Guidance can be evidence-based without putting a real name on a non-clinical app.
Draft notes are the other teaching surface. New providers often know the template and not why a risk line, a diagnosis, or a med rationale has to be there. A live review that flags a missing safety assessment, a thin diagnostic code, a missing reason for the drug, or language that will bounce a claim turns the note into a lesson instead of a checkbox.
Pattern recognition is what experience is. A longitudinal summary (response, relapse signs, how a diagnosis drifted) gives a trainee the movie, not one still.
In a high-volume shop, the same system can flag urgent cases, track note quality across trainees, and give a first pass before the formal hour. The attending still decides. Differentials, safety flags, and options are suggestions. AI does not dictate the plan. That boundary is the product.
## What the research shows
Clinical supervision was associated with improved care processes in most studies included in a 2017 systematic review. Twelve of 14 studies reported improved care processes. Three of six studies reported improved patient health outcomes, including two mental health studies that reported lower symptom severity. ([Snowdon et al., 2017](https://bmchealthservres.biomedcentral.com/articles/10.1186/s12913-017-2739-5))
A 2026 observational study followed 357 youths treated by 108 clinicians at 19 community mental health clinics. Higher levels of data-based feedback and skill rehearsal during supervision were associated with faster and larger improvements in caregiver-rated mental health. Improvement was 1.35 times greater with high levels of data-based feedback and 1.21 times greater with high levels of skill rehearsal. ([Community mental health study](https://pubmed.ncbi.nlm.nih.gov/42387286/))
These studies evaluated human clinical supervision. They did not test AI-delivered supervision. Nextvisit organizes case details and flags documentation gaps for review. The supervisor remains responsible for clinical guidance.
================================================================================
URL: https://nextvisit.ai/blog/ai-medication-tracking-treatment-outcomes-prescribing-patterns-mental-health
Title: Medication history a busy clinic cannot keep in one head
Description: AI can reconstruct medication history, side-effect patterns, and practice-level prescribing that a busy psychiatry clinic cannot keep in one head.
Date: 2026-05-08
Author: Faisal Rafiq, MD
Category: Product
================================================================================
Medication management is where a psychiatry panel outgrows memory. Dose changes, failed trials, side effects, adherence, brand versus generic, augmentation, and new options stack up across hundreds of patients. AI is useful when it puts that history, the side-effect pattern, and the practice's own prescribing data on one screen.
## History the EHR will not assemble
A conventional EHR makes you scroll months of notes and rebuild the pharmacologic story in your head. AI can collapse that into one chronology: start and stop dates, dose changes with the reason, why a drug was stopped, side effects mentioned across visits, what the patient said helped, missed refills, duplicates or conflicts, and past augmentation pairs with the outcome.
"Modern AI systems don't just transcribe what happens in sessions anymore. They structure notes, pull relevant patient history, flag medication interactions, and format everything according to insurance requirements."
## Side effects that only show up across visits
Patients mention fatigue in one visit, sleep weeks later, weight months after that. Isolated complaints do not look like a pattern unless someone lines them up.
AI can connect those points: fatigue after a dose increase, activation or insomnia after an antidepressant start, weight on a given antipsychotic, akathisia or tremor after a start, GI change with a switch, irritability on a stimulant, sexual side effects across several trials. That is faster than rereading the chart by hand, and it is how you change a drug before the patient quits it.
## Withdrawal and abrupt stops
People stop between visits: cost, side effects, a sense that it "wasn't doing anything," or a taper they invented. The clusters to watch are an anxiety spike after a change, flu-like symptoms after a stop, sleep disruption, "brain zaps" after a serotonergic drug, mood relapse, and rapid lability. Catching that early is the difference between a phone intervention and a destabilized week.
## Practice-level prescribing
Individual habits are invisible until someone counts them. AI can show which drugs the practice starts most, which ones get escalated versus stopped, which combinations tend to work or fail, class overuse or underuse, clinician-to-clinician variation, common augmentation, and time-to-response by class. That used to require an academic research shop. It does not anymore.
## Combinations in real charts
Most psychiatric care is combination care, and the trial literature is thin on the pairs you actually use. A practice can look at its own outcomes: an SSRI plus an atypical, mood-stabilizer pairs, a stimulant plus an antidepressant in ADHD with depression, augmentation with buspirone, lithium, or mirtazapine, and what happened when TMS or ketamine sat on top of meds, including after an insurance-mandated switch.
## Brand versus generic
Patients report differences after a substitution. AI can track symptom course, new side effects, relapse, time to improvement, discontinuation, and requests to go back to brand. That is what you take to a payer or a PBM when the chart, not a feeling, is the argument.
## Who might be eligible for the next step
Treatment-resistant depression with several failed antidepressants, weak augmentation, and ongoing SI is the pattern that should raise esketamine. Poor response across antidepressant classes, or bipolar depression that meds have not touched, is the pattern that should raise TMS. The point is to see the candidacy while the patient is still in outpatient care.
## What leadership can see
Directors get a practice-level view: which meds consistently help, where side effects stall treatment, which diagnoses stay resistant, how long stabilization usually takes, how many patients might qualify for TMS or esketamine, and which classes have ugly stop rates. That is quality work you can act on, not a year-end guess.
Early-career clinicians get a history and a comparison set they have not had time to build. Experienced clinicians get the same file without the archaeology: cumulative side-effect burden, outcomes in similar patients, and a ranked list of next options. The prescription is still theirs.
================================================================================
URL: https://nextvisit.ai/blog/ai-mental-health-insurance-audits-documentation-compliance
Title: What auditors look for in a behavioral health note
Description: Payers want medical necessity, add-on psychotherapy content, scales, and matching CPT codes. AI can catch those gaps before you sign.
Date: 2026-05-08
Author: Faisal Rafiq, MD
Category: Compliance
================================================================================
Insurance audits are now a survival issue for mental health practices. Payers want precise notes, accurate CPT codes, and a through-line of medical necessity. High-volume clinics feel that as a second job. AI is useful here when it finds the missing piece before the note is signed.
## What auditors are asking for
Payers want a written reason for med changes, psychotherapy content when you bill 90833 or 90836, and time that matches the session. They also want scales in the chart (PHQ-9, GAD-7, PCL-5, Y-BOCS), a treatment plan with measurable goals, and consistent SI, violence, and safety documentation. Functional impairment and a story that holds up across visits are not optional. Miss one of those and you get a denial, a recoupment, or a clawback.
"The gap between what insurance companies demand and what busy clinicians can realistically document has become a genuine threat to practice viability."
## Catch the gap before the note is final
The practical use is a check on the draft: medical necessity, psychotherapy content that matches the add-on code, a reason for the med change, risk language, function, and a plan that still matches the problem. You fix it in the same sitting. You do not discover it in an audit letter.
## CPT codes
Wrong codes are a main source of mental health denials. The code has to match complexity, clock time, and what the note actually contains. AI can suggest a code from time spent, check that medical decision-making supports the level, confirm add-on psychotherapy criteria, and warn when the billed code is richer than the note.
## Longitudinal necessity
Payers increasingly want the story across visits, not one isolated SOAP. A compiled history (trials and responses, scale trends, hospitalizations, diagnosis changes, adherence, progress toward goals) is what you attach to an appeal.
## Treatment plans
Vague or stale plans fail audits. Measurable goals tied to a diagnosis, interventions that match the modality you billed, updated frequency, and a reason this still needs a licensed clinician are the pieces AI can keep in view as symptoms change.
## Risk, every time it belongs
Psychiatric audits read the risk section closely. SI, HI, self-harm, psychosis or command hallucinations, substance risk, protective factors, and what you did in a crisis need to show up in the notes where they belong, in the same language you would use if someone reread the chart next year.
Practices that put these checks in the workflow see fewer denials, cleaner CPT use, stronger medical-necessity language, faster audit packets, less staff time assembling a response, and fewer clawbacks. That is the job: a note you can defend without adding an hour to the day.
================================================================================
URL: https://nextvisit.ai/blog/ai-transforming-patient-letters-forms-nextvisit-integration
Title: ESA letters, FMLA, and prior auths from the chart
Description: Nextvisit drafts ESA letters, FMLA packets, and prior authorizations from the chart for clinician review.
Date: 2026-09-03
Author: Faisal Rafiq, MD
Category: Product
================================================================================
A patient needs an emotional support animal letter by tomorrow. Another needs FMLA paperwork. Three more are waiting on disability forms. Those documents are real clinical work, and they eat hours. Nextvisit drafts them from the chart: diagnosis, course, and functional language already in the record. It is a HIPAA-bound clinical tool, not a consumer chatbot pointed at a letterhead.
## The paperwork that is not optional
ESA letters, disability evaluations, prior authorizations, and return-to-work notes need clinical information from the chart.
## Measured time savings
The 2024 CAQH Index estimated that fully electronic prior authorization could save medical providers and staff 14 minutes per transaction. ([2024 CAQH Index](https://www.caqh.org/hubfs/Index/2024%20Index%20Report/CAQH_IndexReport_2024_FINAL.pdf))
A Duke Health quality-improvement project reported an 18.5% reduction in EHR time and savings of 1.5 to 6.5 minutes per nursing reassessment. ([Lindsay et al., 2022](https://pmc.ncbi.nlm.nih.gov/articles/PMC9300261/))
The CAQH figure is an industry estimate. The Duke project measured a nursing reassessment workflow. Neither evaluated Nextvisit.
## Draft from the chart, not a blank form
When a clinician asks for an ESA letter based on an anxiety diagnosis, Nextvisit pulls from the patient's timeline: diagnosis codes, treatment duration, symptom severity, and functional impairments. The draft is a formatted letter with the justification already in it.
Nextvisit drafts FMLA packets and prior authorizations from chart information. Review the draft. Correct errors. Send the document.
## The details templates miss
A return-to-work letter has to account for more than the primary diagnosis. Medication side effects, therapy progress, workplace triggers named in session, and accommodations that might make a return stick all belong in the draft.
That used to mean rereading several encounters. The draft now arrives with those pieces assembled, and it flags missing facts or contradictions before you send it.
## Security
Nextvisit encrypts patient data at rest with AES-256 and in transit with TLS 1.2 or later. Role-based permissions control access. Users can enable multi-factor authentication. The audit log records access to patient data.
================================================================================
URL: https://nextvisit.ai/blog/how-ai-is-changing-psychiatry-and-therapy-beyond-the-hype
Title: How AI supports psychiatry and therapy
Description: AI supports structured documentation, longitudinal review, and insurance formatting. Clinicians use this information to make clinical decisions.
Date: 2026-05-08
Author: Faisal Rafiq, MD
Category: Operations
================================================================================
AI can structure documentation, review longitudinal records, and format documents for insurance requirements. Clinicians use these results to make clinical decisions.
## The documentation problem
Documentation competes with time for patient care. When clinicians delay notes, they must reconstruct visit details later.
Transcription captures speech. Clinical documentation organizes relevant visit information in the note.
In psychiatry that means the questions that get lost in a busy panel: when lithium last changed, how many depressive episodes in three years, whether they mentioned sleep last visit.
In therapy it means a psychotherapy note that can satisfy a payer and still contain the work, without an hour of reconstruction after each session.
## Diagnosis and treatment planning
Psychiatry is pattern recognition across time. Anxiety for years before anyone names the hypomanic stretches. Side effects that only look obvious once you line up six visits.
AI can review years of notes quickly and surface:
* Sleep changes that tend to precede mood episodes
* Medication combinations that commonly cause a given side effect
* Past treatment responses that should inform this visit
* Early signs of relapse
The psychiatrist still decides. The file in front of them is less of a blur.
## What therapists are using
Therapy was slower to adopt this, in part because the work feels subjective. The uses that have stuck are the ones that leave the relationship alone:
* Session prep: a summary of the last visits, open themes, and the treatment-plan goals
* Scales and method: PHQ-9 over time, cognitive distortions in the patient's language, next steps on an exposure hierarchy
* Insurance: treatment-plan renewals, medical necessity language, notes formatted for the payer
* Supervision: session summaries a supervisor can actually read
The goal is less cognitive load and less admin, not automated therapy.
## Team care
A psychiatric patient often has a prescriber, a therapist, and a primary care clinician. Coordination is usually phone tag.
A shared clinical summary lets the psychiatrist see what came up in therapy, the therapist see the last med change, and the PCP see the mental health context for a physical complaint. That is ordinary care coordination, done on purpose.
## Training
Residents cannot see every case in person. Structured case reviews, documentation critique at scale, and missed elements turned into teaching points are how volume happens without inventing patients.
## Privacy and ethics
None of this works if patients do not trust it. A system used in psychiatry has to be:
* HIPAA compliant, with encryption and access controls
* Clear about how data is used
* Optional for patients who want a typed note
* Auditable, so a clinician can correct the output
* Subordinate to clinical judgment: the tool suggests, the clinician decides
The line is autonomy. The moment a system decides care on its own, it has stopped being a documentation tool.
The file you can actually use in the room is already useful: ten years of medication trials with what worked, what caused side effects, and which dosing pattern held. A therapist can get a prompt when language suggests suicidal ideation or an avoidance pattern is repeating. The tool extends memory and pattern recognition. It does not replace either.
================================================================================
URL: https://nextvisit.ai/blog/revolutionize-behavioral-health-documentation-with-ai
Title: Notes that finish during the visit
Description: Nextvisit reduces provider administrative time by 68% on average, drafts structured notes, suggests codes, and exports to the EHR.
Date: 2026-09-03
Author: Ryan Yannelli
Category: Documentation
================================================================================
## Real-time documentation
Behavioral health clinicians still lose a large share of the day to notes that do not help the person in the room. Nextvisit listens during the session and drafts a structured note while the encounter is happening.
The draft comes from the visit, not from what you can still remember after clinic. You stay with the patient instead of splitting attention with the keyboard.
Nextvisit reduces provider administrative time by 68% on average, based on Nextvisit customer surveys and product metrics from April 2023 to June 2026.
## Billing accuracy
Billing errors in this specialty turn into denials. Nextvisit suggests billing codes from the structured note it created in the session, so the code is tied to what was documented, not to what you hoped you wrote.
The same record is what you hand an auditor: what happened, what you did, and what you billed.
## EHR integration
Nextvisit does not replace the EHR. It exports the structured note into the chart you already use. You are not learning a second system of record. Clinicians adapt because the tool sits in the visit they already run.
## Security and privacy
Nextvisit is built for HIPAA. Data is encrypted in transit and at rest. Access is controlled. Every touch of patient data leaves an audit trail.
## Built for behavioral health
Nextvisit is trained on psychiatric and therapy sessions: the language, the assessment frames, and the treatment work those visits contain. The draft is a behavioral health note, not a generic medical note with the specialty swapped in.
================================================================================
URL: https://nextvisit.ai/blog/automated-progress-notes-streamlining-medical-documentation
Title: Progress notes that write during the visit
Description: Automated progress notes capture the encounter as it happens and write a structured record into the EHR, so the chart is not rebuilt after clinic.
Date: 2026-05-08
Author: Ryan Yannelli
Category: Documentation
================================================================================
## Why notes get automated
Manual progress notes are slow, and they are often incomplete by the time the day ends. Volume goes up. Payer rules do not get shorter. The gap between what has to be in the chart and what a clinician can type after the last visit keeps widening.
Automated progress notes capture the encounter as it happens and turn it into a structured record. The point is a complete note without a second shift at the keyboard.
## What changes in the day
The tool writes during the visit, so you are not reconstructing the hour from memory. Attention stays on the patient.
When the tool writes into the EHR you already use, you are not re-entering the same facts in a second window. Notes also look more alike across providers, which is what covering clinicians and billers need.
## Accuracy, and the evening
A draft written from the visit misses less than a note written at 7pm. After-hours documentation is a common reason people leave a practice. Finishing during or right after the encounter gives the evening back.
## What people worry about
Workflow disruption is the usual objection. Tools that sit next to the current EHR, rather than replacing it, keep the learning curve short.
The system uses encryption, access controls, and audit trails. Regulated organizations must meet applicable HIPAA and state requirements.
Behavioral-health tools are already past plain transcription. They structure the visit and surface patterns in the same workflow.
# Guides
================================================================================
URL: https://nextvisit.ai/guides/ariamd-voice-week-one
Title: Setting up your AriaMD voice in week one
Description: A five-day checklist: what to correct, what to template, and the settings that stop the same edit from repeating.
Date: 2026-09-03
Audience: all
Reading time: 3 min
================================================================================
Use this checklist to tune AriaMD for routine visit types. Correct repeated errors during the week.
## Day 1: just chart the way you would
Do not optimize on day one. Run Aria on every visit. Edit drafts the way you would edit any draft. The system records the diff between what it produced and what you signed. That diff is the cleanest signal of what you actually wanted.
End-of-day checklist:
* Did you run Aria on every routine visit? (Yes is the goal.)
* Did you edit drafts to match how you actually chart? (Yes.)
* Were any drafts so far off that you discarded them and started over? (If yes, note the visit type. Day three is for that.)
Time spent: zero extra minutes. Chart normally.
## Day 2: notice what you are editing repeatedly
After a second day, the same change shows up two or three times. Examples from real users:
* "Changing 'patient denies suicidal ideation' to 'no SI reported' every time."
* "Moving the medication list from inside the HPI to its own line."
* "Reformatting the MSE from a paragraph to bullet points."
* "Changing 'follow-up in 4 weeks' to 'RTC 4 weeks.'"
Write the patterns down. Three to five repeated edits is typical.
End-of-day checklist:
* List the 3 to 5 corrections you made repeatedly today.
* Decide which are stylistic (the system can learn these) vs. structural (better as a template setting).
Time spent: 5 minutes at end of day.
## Day 3: turn structural preferences into settings
Open Nextvisit settings. The structural levers are there: bullet vs paragraph for MSE, "no reported" vs "patient denies," default follow-up format, and whether the assessment is a numbered list or prose.
Set these once. Recurring structural corrections leave the edit pass.
End-of-day checklist:
* Style preferences updated to match how you chart.
* Edits today decreased compared to days 1 and 2. (Should be visible.)
* Recurring corrections that are not settings: those are stylistic, and the system is learning them.
Time spent: 10 minutes.
## Day 4: build your first custom template
Most clinicians have one or two visit types the defaults miss: a structured new-patient intake, a return visit with a fixed question set, a med-management follow-up with a fixed format.
Pick one. Open the template editor, paste an anonymized sample of how you actually write that visit, and convert it into a reusable template. Apply it to the next two or three visits of that type. The third pass should be close to your final note.
End-of-day checklist:
* One custom template imported and refined.
* That template applied to at least two visits.
* Edits on those visits trending toward minimal.
Time spent: 15 minutes for setup, then time saved on every later visit of that type.
## Day 5: review what you have and tune
Open the personalization summary in settings (Nextvisit shows active style preferences in plain English). Confirm they match how you chart. Edit anything that does not. Reset and retrain anything that went the wrong way.
End-of-day checklist:
* Active style preferences match how you chart.
* You have at least one custom template running.
* Editing time per draft is meaningfully lower than day 1.
* Anything still consistently wrong is in a support ticket or a question for your account team.
Time spent: 10 minutes.
## Administrative time
AriaMD reduces provider administrative time by 70%, based on Nextvisit customer surveys and product metrics from April 2023 to June 2026.
## What week 2 looks like
Continue to tune custom templates for visit types that the defaults miss.
================================================================================
URL: https://nextvisit.ai/guides/complex-psych-medication-management-visit
Title: Documenting a complex outpatient psychiatric medication management visit
Description: How to chart a polypharmacy, partial-response med-management visit: interval history, regimen list, assessment reasoning, and plan.
Date: 2026-05-03
Audience: psychiatrist, np-pa
Reading time: 4 min
================================================================================
The visit is familiar: multi-year history, three or four active diagnoses, four to seven medications, partial response, a new stressor, and a decision to escalate, augment, switch, or hold. The note is harder. Standard SOAP does not hold the chain of reasoning, so the assessment shrinks to two sentences and the HPI becomes six paragraphs of medication history that belongs somewhere else.
## What the visit actually is
A complex med-management follow-up has six questions:
1. How is the patient doing now, in their words and on mental status?
2. What has changed since last visit: symptoms, function, life context, adherence?
3. What is the current regimen: doses, durations, tolerability?
4. What is the response and tolerability picture across the regimen?
5. What is the clinical decision today, and why?
6. What is the plan and the reassessment timeline?
The note should make all six visible. A tight assessment that ties the answers together is the part the chart hinges on.
## Section by section
### Chief complaint
One sentence, close to the patient's language. "Follow-up on bipolar II, depressive episode, with concern about a flat plateau in mood since the last visit" is a chief complaint. "Follow-up" is not.
### Interval history (not HPI in the classic sense)
This is where most complex med-management visits dump too much content.
What belongs:
* Symptom trajectory since last visit, with patient-reported severity and any objective markers (PHQ-9, GAD-7, MDQ, sleep tracking, side-effect logs).
* Medication adherence, with corroboration where relevant (refill history, blood levels for lithium or valproate, UDS for controlled substances).
* Functional status: work, school, relationships, sleep, appetite, exercise, substance use.
* New stressors or context changes.
* Side effects, in the patient's terms.
* Acute events: ED visits, hospitalizations, incidents.
What does not belong:
* Full medication history. Put it in the medication list if it is not already in the longitudinal chart.
* Past psychiatric diagnoses. Those belong in past psychiatric history.
* Detailed family history, except where it changes today's decision.
* The differential. That belongs in the assessment.
A complex interval history should fit in three to five clear paragraphs. If it sprawls, content is in the wrong section.
### Mental status examination
The MSE is where the chart shows you were in the room. Make it specific. Identical MSE paragraphs across visits are the highest-yield target in a chart review.
What "specific" looks like:
* Affect in observed terms with range and reactivity ("constricted, mildly depressed, reactive within a narrow range") rather than stock adjectives.
* Thought process in observed terms ("linear and goal-directed today, in contrast to the loose associations noted at the prior visit").
* Speech, psychomotor activity, and any cognitive observations from the conversation.
* Suicide and homicide risk in the form actually assessed, with reference to the C-SSRS or your standard instrument.
* Insight and judgment in clinical terms, not boilerplate.
If the MSE is word-for-word the prior visit and the rest of the note describes a change, that inconsistency is a vulnerability. Edit the MSE to match the visit.
### Current regimen and tolerability
Render the regimen as a structured list, not prose. Each medication: drug name, dose, frequency, duration on current dose, indication, and tolerability.
```
Lamotrigine 200 mg HS, on this dose since 2024-07, indicated for bipolar II maintenance. Tolerated well, no rash, no cognitive complaints.
Bupropion XL 300 mg AM, on this dose since 2025-11, indicated for bipolar II depression augmentation. Mild jitteriness initially, resolved.
Quetiapine 50 mg HS, on this dose since 2026-01, indicated for sleep and anxiolysis. Mild morning grogginess, manageable.
Hydroxyzine 25 mg HS PRN, used 2-3x/week, indicated for anxiety PRN. No issues.
```
Four lines, not four paragraphs. A covering colleague or a reviewer can scan it.
### Response and tolerability across the regimen
A short paragraph that ties symptoms to the current regimen.
> Patient is in partial remission on the current regimen. Depressive symptoms are improved from a PHQ-9 of 19 in 2025-11 to 12 today, with most of the improvement in the first 8 weeks of bupropion augmentation. Residual symptoms are concentrated in sleep, energy, and motivation. Anxiety is improved from a GAD-7 of 16 to 9, with quetiapine providing the most subjective benefit at the cost of mild morning grogginess. Tolerability is acceptable across the regimen. No metabolic, hepatic, or cognitive concerns on the current monitoring.
That paragraph does the work a templated assessment skips. It connects the patient's experience to the medications.
### Assessment
Answer two questions: what is the formulation today, and what is the decision.
> Bipolar II disorder, currently in a partial-response depressive episode of approximately 6 months' duration, with anxiety as a long-standing comorbidity. The current regimen has produced meaningful but incomplete response. The depressive plateau in the past 4 weeks, in the context of stable adherence and adequate trial duration on the current doses, suggests the regimen has reached its asymptote. The clinical decision today is between augmentation with a third agent, rotation of the bupropion to another stimulating antidepressant, or addition of psychotherapy intensification rather than further pharmacologic change. Patient prefers a non-pharmacologic addition first; agreed.
Everything before this sets up the reasoning. Everything after it implements the decision.
### Plan
Render by domain:
* Medications: continue current regimen at current doses for 4 more weeks. No changes today.
* Psychotherapy: increase from biweekly to weekly with the current therapist. Discussed coordination of care.
* Monitoring: PHQ-9 and GAD-7 at next visit. Repeat lithium level not indicated this cycle.
* Safety: C-SSRS administered today, low acute risk, low to moderate chronic. Standard safety plan in effect. Crisis line reviewed.
* Follow-up: 4 weeks. Sooner if symptoms worsen.
Make the plan specific, time-bound, and tied to the assessment.
## The decision-justification problem
Payer audits and peer reviews focus on the decision. Make the reasoning easy to find.
If you continue the regimen, say why. "Adequate trial duration on the current dose, partial response, patient preference for non-pharmacologic intensification first" is reasoning. "Continue current treatment plan" is not.
If you escalate, connect symptoms, trial duration, and the next step. "Inadequate response after 12 weeks on bupropion 300 mg with full adherence; rotating to vortioxetine 10 mg daily to assess for differential response in the depressive presentation" is reasoning.
If you taper or stop a medication, show the rationale and the monitoring plan for the change.
Same pattern in all three: connect the data to the decision in prose, in the assessment, not as a bullet list.
## Where Nextvisit fits
AriaMD already has the chart, prior notes, medication history, and scale trends, so today's interval history can show what changed instead of restating the whole history. The medication list pulls from `/patient/[uuid]/medications`. Treatment Pulse and AI Timeline keep prior responses and decisions visible without paging through notes. Put the clinical-reasoning ask in the med-management template; the custom-template guide at `/guides` covers the editor.
================================================================================
URL: https://nextvisit.ai/guides/spravato-session-documentation
Title: Documenting a Spravato session step-by-step
Description: REMS-aware Spravato documentation: pre-session checks, vitals timepoints, observation, discharge criteria, and a sample template.
Date: 2026-04-30
Audience: psychiatrist, np-pa
Reading time: 3 min
================================================================================
A Spravato visit takes two hours: REMS attestations, vitals at multiple points, and a 2-hour observation period. Payers audit these claims more often than almost any other psychiatric service. Documentation defects are the usual denial.
A working template is in the library at `/templates`.
## Pre-session
The pre-session block establishes that the patient is eligible for treatment today. Document four things.
REMS attestation: the patient is enrolled in the Spravato REMS, the prescriber is REMS-certified, and the dispensing pharmacy is REMS-certified. Most practices have this as a checkbox. It should also appear in the note text so it is searchable on audit.
Indication: treatment-resistant depression with adequate trial documentation, or major depressive disorder with acute suicidal ideation per the FDA label. Reference the clinical assessment that supports the indication. Do not only state it.
Current medications: full active list at session start, including any antidepressant the patient is required to be on per the indication, with dose and adherence.
Baseline vitals: blood pressure, heart rate, and any pre-session screening measures (PHQ-9, C-SSRS) administered that day.
## Administration
Dose, route, and tolerance during administration. Dose (56 mg or 84 mg, with the device count), route (intranasal), administration time, and the patient's response in the first 10 minutes (alert, oriented, comfortable, no acute adverse events). If you observe dissociation, sedation, or a vital-sign change above your protocol threshold, that goes here.
## Observation period
The 2-hour observation period is where most documentation defects appear. The note should show the patient was observed for the full 2 hours by qualified personnel, with vitals and clinical status at the intervals your protocol specifies (typically 0, 40 min, 60 min, 90 min, and 120 min).
For each timepoint: blood pressure, heart rate, level of consciousness, presence or absence of dissociation, presence or absence of significant adverse effects. If there was a transient effect (mild dissociation, mild blood-pressure elevation that resolved), document it and document the resolution. Charts that say "no adverse events" when one timepoint shows a BP of 162/98 are the ones that get bounced.
## Discharge criteria
Before the patient leaves, document that they meet discharge criteria: vitals stable and in an acceptable range, fully alert, no acute adverse effects, able to follow safety instructions about driving and operating machinery for the rest of the day, ride home arranged, follow-up scheduled.
If they do not meet criteria at 2 hours, they stay. Chart the extended observation, the reasoning, and the eventual discharge.
## Assessment and plan
The assessment summarizes today's response and the trajectory across the protocol if this is not the first session. PHQ-9 trend across sessions belongs here. The plan covers the next session date, the next dose, any adjustments to concomitant medications, and the safety plan in effect until the next visit.
## Sample template structure
```
Patient: [name, DOB, MRN]
Session #: [n] of induction or maintenance phase
Date and time: [start to end]
Pre-session
- REMS attestation: prescriber, pharmacy, patient enrollment
- Indication: TRD, adequate antidepressant trials documented
- Current medications: [list]
- Baseline: BP, HR, PHQ-9, C-SSRS
Administration
- Dose: 56 mg / 84 mg (n devices)
- Route: intranasal
- Time of administration: [time]
- Initial response: [10-min check]
Observation
- T+0: BP / HR / LOC / dissociation / AE
- T+40: BP / HR / LOC / dissociation / AE
- T+60: BP / HR / LOC / dissociation / AE
- T+90: BP / HR / LOC / dissociation / AE
- T+120: BP / HR / LOC / dissociation / AE
Discharge
- Vitals at discharge
- Mental status at discharge
- Discharge criteria met
- Ride home confirmed
- Next visit scheduled
Assessment
- Response to today's session
- PHQ-9 trend across sessions
- Tolerability
Plan
- Next session date and dose
- Concomitant medication
- Safety plan in effect
```
## Where Nextvisit fits
Aria captures the conversational portions (pre-session check-in, post-administration discussion, discharge instructions) and routes them into the right sections. Vitals timepoints come in as structured data from your tablet or paper form, and the system fills the table. The assessment drafts from the captured conversation and score trends. You review and sign.
The full Spravato template is at `/templates`. Customize it for your vitals interval and protocol-specific add-ons.
================================================================================
URL: https://nextvisit.ai/guides/mat-note-requirements-2026
Title: MAT note requirements 2026
Description: What 2026 MAT notes need for buprenorphine, methadone, and naltrexone: intake fields, per-visit items, UDS, and 42 CFR Part 2 handling.
Date: 2026-04-18
Audience: psychiatrist, np-pa
Reading time: 4 min
================================================================================
MAT notes sit under 42 CFR Part 2, plus SAMHSA, DEA, state, and payer rules. This is what the 2026 note has to contain for buprenorphine, methadone, and naltrexone.
## Initial intake
The initial MAT intake is longer than a routine visit because it has to establish medical necessity, document informed consent, capture the substance-use history that supports the diagnosis, and start the safety and monitoring plan.
Required sections at intake:
* Full substance-use history (substances, age of first use, current use pattern, prior treatment history, periods of remission).
* Medical history relevant to MAT (liver function, cardiac history, pregnancy status, current medications including controlled substances).
* Mental health history including suicidality, self-harm, and current psychiatric diagnoses.
* DSM-5 substance use disorder diagnosis with severity specifier.
* Current functional status (housing, employment, family, legal).
* Current symptom status (withdrawal symptoms today, COWS score for opioid users, CIWA for alcohol).
* Treatment plan including induction protocol, follow-up cadence, monitoring plan, and counseling referral.
* Informed consent: medication risks and benefits, alternatives discussed, treatment agreement signed.
* Baseline UDS result.
* Naloxone prescription documented (for opioid MAT).
## Buprenorphine specifics
Buprenorphine has the most flexible federal framework after the X-waiver removal in 2023. Keep the documentation standard rigorous. Audits are common.
* Each visit: dose, fills since last visit, evidence of taking the medication as prescribed (UDS or self-report with corroboration), withdrawal/craving status, side effects, functional status, and counseling participation.
* Dose escalations require a documented clinical rationale.
* Diversion concerns (early refill requests, missed appointments, UDS results inconsistent with the prescription) require a documented response: counseling discussion, increased visit frequency, supervised dosing if applicable, or referral to a higher level of care.
* Ongoing UDS at a frequency consistent with patient stability (typically monthly in the first three months, then quarterly in stable patients, more often if there is concern).
## Methadone specifics
Methadone for opioid use disorder is regulated at the federal OTP (opioid treatment program) level. The documentation is more prescriptive.
* Initial induction: state rules vary. Typical practice is supervised dosing with daily reassessment for the first weeks. Each daily dose visit needs dose, observed administration, baseline vitals if first dose, and clinical status.
* Take-home doses: each take-home requires documented eligibility against the SAMHSA take-home criteria (8 axes: absence of recent abuse, regularity of attendance, absence of behavioral problems, absence of recent criminal activity, stability of home environment, length in treatment, assurance of safe storage, rehabilitative benefit).
* Annual review: annual documentation of treatment plan review, justification for ongoing maintenance, and any updates.
## Naltrexone (extended-release injectable) specifics
Naltrexone is less heavily regulated. The documentation still has a pattern.
* Each injection visit: confirmation of opioid abstinence (typically 7+ days for opioid use disorder, naloxone challenge or UDS as protocol indicates), absence of acute opioid intoxication, injection site, dose, and observation period.
* Liver function monitoring at baseline and periodically.
* Documented counseling about the risk of overdose if relapse occurs after the dose wears off.
## Per-visit ongoing documentation
For all MAT regimens, the ongoing visit note needs:
* Substance use since last visit (with corroborating UDS).
* Medication adherence.
* Withdrawal/craving status (use validated scales when applicable).
* Functional status in 4 domains: physical, psychiatric, social, occupational.
* Counseling participation (or documentation of referral and outcome).
* Plan: continue, adjust, or escalate.
## 42 CFR Part 2 considerations
MAT records are SUD records under 42 CFR Part 2. The documentation does not change. The handling of records does. Disclosure to outside parties requires patient consent in the Part 2 format, even within the same health system in many cases. The chart should show that the patient was informed of the confidentiality protections and any limits.
## Sample assessment language for ongoing visits
A defensible MAT assessment for an ongoing visit:
> Patient is an N-year-old [demographics] with opioid use disorder, severe, in [early/sustained] remission on [medication and dose]. At today's visit, patient reports [no/minimal/moderate/severe] cravings, [no/some] withdrawal symptoms, and [no use/use] since last visit. UDS today is [result], consistent/inconsistent with prescription. Functional status [stable/improving/worsening] in [domains]. Counseling [participation summary]. Plan: continue current dose, follow-up in [interval], next UDS [interval].
The Nextvisit MAT template at `/templates` covers this format and the buprenorphine, methadone, and naltrexone variants. Dose, UDS, and COWS or CIWA land as data, not HPI prose. Missing required elements show up before you sign.
================================================================================
URL: https://nextvisit.ai/guides/build-custom-intake-template
Title: Build a custom intake template in 15 minutes
Description: Build a Nextvisit intake template: editor path, variable tokens, required sections, and when to split into a second template.
Date: 2026-04-12
Audience: all
Reading time: 4 min
================================================================================
A psychiatrist's intake does not read like a therapist's, which does not read like an NP's medication-management intake. Default templates are a starting point. The same pattern works for any visit type.
The template editor is at `/apps/custom-prompts`. Editing an existing template is at `/apps/custom-prompts/[id]/edit`.
## Step 1: bring a real intake note
Open a recent intake you wrote yourself, ideally one you would be happy to receive from a colleague. Anonymize it. Strip the PHI. Keep the structure, section headings, prose patterns, and level of detail.
The editor mirrors the example you give it. A polished real note in, a polished template out.
If you do not have a recent intake that meets the bar, write one against a hypothetical patient with real clinical detail. Ten minutes on the example is the difference between a template you keep and one you replace next month.
## Step 2: open the template editor
Go to `/apps/custom-prompts` and click New Template. Name it so it is obvious in six months. "Intake (psychiatry, adult)" beats "intake_v3."
Two areas matter: the prompt body (what you want the note to look like) and the variable token list (what AriaMD substitutes at runtime).
## Step 3: paste the example, then describe the structure
Start the prompt body with one or two sentences on visit type and audience:
> Adult outpatient psychiatry intake, 60 minutes. Generate the note in the structure below, in the voice of the treating psychiatrist, using clinical vocabulary appropriate to outpatient adult psychiatry.
Paste the anonymized example below it. Above the example, write "Use the following note as the structural and stylistic reference."
The prompt sets intent. The example sets shape. Together they beat either one alone.
## Step 4: insert variable tokens
Replace demographic and visit-context fields with tokens. Available at the time of writing:
* `[patient_name]`
* `[dob]`
* `[age]`
* `[gender]`
* `[profession]`
* `[provider_name]`
* `[duration]`
A header like "Jane Doe, 34F, occupation: librarian, seen by Dr. Smith on 4/12/2026" becomes "[patient_name], [age][gender], occupation: [profession], seen by [provider_name]." AriaMD substitutes the real values at runtime.
Do not over-tokenize. Tokens are for fields that come from the patient record. Everything else is generated from the encounter audio and your prompt.
## Step 5: structural levers worth setting explicitly
Spell these out in the prompt:
* MSE format. "Render the mental status exam as a single paragraph in prose, not as a bullet list" or the reverse.
* Diagnosis style. "Render diagnoses with ICD-10 codes in parentheses after the diagnostic name. Order primary first."
* Assessment voice. "Write the assessment in clinical reasoning prose that ties findings to the diagnosis and the plan. Avoid bulleted differential lists unless the case calls for it."
* Plan structure. "Organize the plan by domain: medications, psychotherapy, monitoring, follow-up, safety. Use short bullet items under each domain."
* Length. "Total note length should target 1.5 to 2.5 pages of clinical content for a standard 60-minute intake. Trim sections where the patient had little to report."
Those five sentences remove most day-one structural editing.
## Step 6: handle the things every intake needs
State the sections that always belong in an outpatient psychiatric intake:
> Always include sections for: chief complaint, history of present illness, psychiatric history, medical history, family psychiatric history, social history, substance use history, current medications, allergies, mental status examination, suicide risk assessment with C-SSRS reference, diagnosis, and plan. If the visit did not produce data for a section, mark it explicitly (e.g., "Family psychiatric history: deferred this visit, will obtain at follow-up") rather than omitting the section.
A marked deferral is more defensible than a missing section. A reviewer can see the field was considered.
## Step 7: save, run, and edit one real visit
Save the template. On your next intake, select it as the note format. Record as usual. Read the draft as you would a colleague's note and edit what does not match how you chart.
The first generated note is rarely perfect. The second is closer. By the third intake, the remaining edits are usually patient-specific content, not structure.
## Step 8: come back and tune
After three to five intakes, open the template again. Compare what you wrote in the prompt to what you signed. Recurring edits are the next prompt update.
Common second-round refinements:
* Adjusting a section that was consistently too long or too short.
* Tightening assessment voice if the prose is more textbook than clinical.
* Adding a sentence for a specific population (e.g., "When the patient endorses substance use, generate a brief substance-use review even if not part of the chief complaint").
* Removing a section that was not actually used.
Two rounds of tuning is typical. After that the template stabilizes.
## When to make a second template instead of forcing one
Do not make one template do adult outpatient, geriatric, child and adolescent, and addiction medicine. If two visit types have different structures, make two templates and pick at the start of the encounter. Templates are cheap. Bad notes are not.
A reasonable starting set for a psychiatric practice: adult intake, adult med-management follow-up, child/adolescent intake, child/adolescent follow-up. A therapist might run: intake, DAP follow-up, BIRP follow-up, treatment-plan review. An NP/PA in addiction medicine might run: MAT intake, MAT follow-up, naltrexone visit, buprenorphine induction.
The template runs on every visit you assign it to. Aria captures the audio, applies the prompt, substitutes the tokens, and produces the draft. The same prompt body can also drive AI Tasks.
================================================================================
URL: https://nextvisit.ai/guides/defensible-psych-notes-ai
Title: Writing defensible psychiatric notes with AI in the room
Description: How to edit an AI draft so it does not read as templated: prune filler, vary phrasing, own the assessment, and attest on sign-off.
Date: 2026-04-08
Audience: psychiatrist, np-pa, therapist
Reading time: 4 min
================================================================================
Repeated structure and stock phrases can make a note appear templated. Review each draft so the note reflects the visit and the clinician's judgment.
## What "AI-shaped" looks like
Three patterns make a note read as generated:
Smooth completeness. Every section is filled at roughly the same length. Real clinicians chart unevenly. The HPI is long when the case is complex; past medical history is short when it is unremarkable.
Stock phrasing. "Patient denies suicidal ideation, homicidal ideation, audio or visual hallucinations" appearing verbatim across every patient. Real clinicians vary the phrasing.
Structural rigidity. Every assessment ends with a numbered list of three differentials, in the same order, with the same connecting logic. Real assessments adapt to the case.
## Edit pattern 1: prune the smooth completeness
Read the draft and ask: where is the chart artificially full? If the patient has no relevant family history, "no contributory family psychiatric history" is enough. You do not need a paragraph on what was reviewed and negative.
If the MSE is a stock paragraph and the visit showed nothing abnormal, leave it short. "MSE within normal limits, alert and oriented x4, mood and affect appropriate to context, no evidence of psychotic symptoms." Three sentences. Real.
Long, formulaic MSE paragraphs on a stable patient mark templated documentation.
## Edit pattern 2: vary phrasing across visits
If the note for visit N+1 is structurally identical to visit N, edit it. Specific differences this visit produced specific words. Find one or two phrases per section that change. The cumulative effect is significant.
Nextvisit's personalization helps here. The system learns your phrasing variations instead of locking onto one stock phrase. The MSE on a stable patient at visit 4 should not match visit 5 word for word.
## Edit pattern 3: own the assessment
Reviewers focus here. It should sound like your clinical reasoning, not a textbook summary.
Concretely:
* Explicit clinical reasoning. "Patient reports decreased symptoms on sertraline 100 mg with adequate trial duration, suggesting partial response. Considering augmentation versus rotation given residual sleep disturbance and hypoarousal."
* Patient-specific context. "Given recent job loss and the timing of symptom onset, situational contribution is significant though does not change the diagnosis."
* Decisions and the reasons for them. "Continuing current dose given functional improvement. Will revisit at 4 weeks if residual symptoms persist."
Avoid: "Patient presents with major depressive disorder, recurrent, moderate. Will continue current treatment plan." That is a template, not an assessment.
## Edit pattern 4: keep the patient's voice
Direct quotes are one of the strongest signals that a clinician was in the room. Aria preserves quotes when they are clinically meaningful. Keep them. Edit only if the patient said something off the record or the quote is too long to be useful. A two-line quote in the chief complaint or a sharp phrase in the HPI is a feature.
## Edit pattern 5: handle uncertainty honestly
If the assessment includes uncertainty, the note should say so. "Differential includes adjustment disorder versus emerging major depressive disorder; will reassess at 2 weeks given recent stressor." That is more defensible than a confident diagnosis the next visit may revise.
## Sourcing
When the chart references a prior medication trial or a prior diagnosis, attribute it. "Per patient report" and "per prior records reviewed" are different attributions. They matter clinically and legally. AI drafts sometimes blur the distinction. Keep the source clear.
## The sign-off attestation
The signature is your representation that the chart reflects the visit and your judgment. We recommend an attestation that names the AI-assisted draft:
> This note was generated as a draft from an AI-assisted ambient documentation system, then reviewed and edited by the undersigned clinician. The clinical content reflects the clinician's judgment and the visit as conducted. The clinician takes full responsibility for the accuracy and completeness of this record.
Check your state medical board guidance on AI documentation. Update the attestation as needed.
Review each draft. Correct errors. Sign the note. Nextvisit flags low-confidence passages and phrases that match recent notes. Use these flags during your review.
================================================================================
URL: https://nextvisit.ai/guides/cssrs-risk-documentation
Title: C-SSRS scoring and risk documentation
Description: When to give the C-SSRS, how to score and chart it, and assessment language that holds up if a visit becomes a sentinel event.
Date: 2026-03-28
Audience: psychiatrist, np-pa, therapist
Reading time: 4 min
================================================================================
The Columbia Suicide Severity Rating Scale (C-SSRS) is the most widely used structured suicide-risk instrument in outpatient behavioral health. It is short, validated across populations, and structured enough that two clinicians scoring the same patient should land nearby. It is also frequently mis-charted: pasted into the HPI as free text, scored without the qualifying questions, or skipped on a stable patient who later has a bad week.
## When to administer
Give the C-SSRS where suicide risk is most likely to change or be missed:
* Every new-patient intake. A negative C-SSRS on day one is the reference point for later visits.
* Any visit where the patient endorses item 9 on the PHQ-9 or mentions suicidal thoughts, plans, or behaviors.
* Any visit where the picture changes: new diagnosis, new psychosocial stressor, recent hospitalization, recent self-harm, recent loss.
* Before and after major treatment changes that carry suicide-risk implications: new antidepressant in adolescents, dose escalation, taper of a stabilizing medication, transition from inpatient or partial hospitalization.
* On a routine cadence for high-risk patients in active treatment, even when the picture looks stable. Quarterly is a reasonable default; some practices use every visit.
Do not give it to everyone at every visit. Used that way it becomes noise.
## Which version
Three versions are in clinical use:
* Full Lifetime/Recent. The complete instrument, for intake and complex cases.
* Since Last Visit. The follow-up version, focused on what changed.
* Screener. A 6-item short form for triage and high-volume settings.
Pick one per visit type and stay consistent. Most outpatient practices use Lifetime/Recent at intake and Since Last Visit at follow-up, with the Screener reserved for crisis triage or low-acuity primary care.
## How to administer
Administer it as an interview, not a paper handout. The qualifying questions ("Have you actually had any thoughts of killing yourself?" before going further) require you to hear the answer and the nuance. Hesitation, a qualifier, or a wording shift is clinical information a Likert form will lose.
Standard order:
1. Wish to be dead (item 1).
2. Non-specific active suicidal thoughts (item 2).
3. Active suicidal ideation with method (item 3).
4. Active suicidal ideation with intent (item 4).
5. Active suicidal ideation with plan and intent (item 5).
6. Behavior in the relevant lookback period (actual attempt, interrupted attempt, aborted attempt, preparatory acts, non-suicidal self-injury).
If item 1 is negative, skip items 2 through 5 per the standard scoring rule. Lethality questions follow if any behavior is endorsed.
Do not change the wording. The validation rests on the wording.
## How to score
Two scores come out. The ideation score (1 through 5) is the highest endorsed ideation level. The behavior score is categorical (no behavior, preparatory, aborted, interrupted, attempt) with a lethality rating when an attempt is endorsed.
Common traps:
* A patient who says "I would never act on it" still endorses the underlying ideation. Item 4 (intent) captures that distinction. Score the ideation honestly and let item 4 do its job.
* Non-suicidal self-injury (NSSI) is captured separately. NSSI without suicidal intent is clinically important and chartable. It is not a suicide attempt.
* The lookback window matters. "Since last visit" is shorter than "in the past month" is shorter than "lifetime." Use the version that matches the window you intend to assess.
## How to chart
Put the C-SSRS in structured fields, not a paragraph in the HPI. Same reasons as PHQ-9 and GAD-7, plus one more: when a patient has a bad outcome, the chart will be reviewed. A structured C-SSRS with date, version, item-level responses, and clinician interpretation is far more defensible than "C-SSRS negative" buried in HPI prose.
A complete charted C-SSRS includes:
* Version administered (Lifetime/Recent, Since Last Visit, Screener) and lookback window.
* Item-level responses for items 1 through 5.
* Behavior section: any preparatory, aborted, interrupted, or actual events in the lookback window, with lethality rating where applicable.
* Most recent behavior date if any behavior was ever endorsed (lifetime).
* Clinician interpretation: low, moderate, or high acute risk; low, moderate, or high chronic risk.
* Reasoning that ties the score to the clinical picture.
* The risk-mitigation plan in effect after the visit.
In Nextvisit, the C-SSRS lives in Scales and Exams on the patient profile (`/patient/[uuid]/scales`) as a structured instrument. Aria captures the conversational portion and routes responses into those fields. The trend shows in the patient header. The risk interpretation belongs in the assessment as prose as well.
If your system does not support a structured C-SSRS, use a dedicated EHR-template section with a consistent format. Free text in the HPI is the worst option for this scale.
## Risk interpretation: the part that matters
The clinician determines the risk level from the C-SSRS score and the clinical context.
A defensible interpretation looks like this:
> Acute suicide risk assessed as moderate. Patient endorses ideation at item 3 level (suicidal thoughts with method, no intent or plan) in the past week, in the context of recent job loss and disrupted sleep. No behavior endorsed in the lookback window. Chronic risk assessed as moderate based on prior attempt in 2023, sustained depressive episode this year, and limited social supports. Protective factors include engaged treatment, stable housing, and access to family. Plan addresses access to means, increased visit frequency, and crisis-line review with the patient.
Reviewers and auditors are looking for the reasoning, not the number.
## What "low acute risk" should not look like
"Patient denies SI/HI. Low risk." on every chart, regardless of context, is templated documentation. If the patient just disclosed a divorce and the assessment matches the prior six visits, the chart will not hold up. Even on a stable patient, "no SI endorsed today; risk remains low in the context of stable medication, engaged treatment, and intact supports" is closer to defensible than the boilerplate.
## When the assessment changes the plan
* Low acute, low chronic: routine follow-up, standard safety planning if any history.
* Moderate acute, low to moderate chronic: increased visit frequency, means restriction, written safety plan, family or support involvement where appropriate.
* High acute or imminent risk: do not let the patient leave without a higher level of care. Document the disposition (ED transfer, voluntary admission, mobile crisis, partial hospitalization) and the handoff.
"Follow up in 2 weeks" is not a plan after a moderate-acute assessment. "Follow up in 1 week, crisis line reviewed, partner aware of the elevated risk and lock box obtained for medications, return precautions discussed" is.
In Nextvisit, the C-SSRS is a structured scale on the patient profile. Aria lands the responses. Your interpretation belongs in the assessment. The trend is on the timeline.
================================================================================
URL: https://nextvisit.ai/guides/phq9-gad7-best-practices
Title: PHQ-9 and GAD-7 scoring and trend documentation
Description: When to give PHQ-9 and GAD-7, how to score and trend them, and how to keep scores structured for MBC reporting.
Date: 2026-03-22
Audience: therapist, np-pa, psychiatrist
Reading time: 3 min
================================================================================
PHQ-9 and GAD-7 are the screeners that show up most often in payer-mandated MBC reporting. They are short, free, and well-validated. They are also captured poorly: scored inconsistently, stuck in the HPI as free text, or never trended.
## When to administer
Give PHQ-9 and GAD-7:
* At every new-patient intake.
* At every follow-up during active treatment of depression or anxiety.
* On a defined cadence (every visit or every other visit) for patients in maintenance.
* Before any major treatment change (medication initiation, dose change, regimen switch, course change).
* When the patient or family raises concern about symptom trajectory.
Most practices land on every visit during active treatment, every other visit during maintenance. That is a reasonable default.
## How to administer
Two patterns work:
Pre-visit, on a tablet or via the patient portal. The patient completes the screener before the visit. Scores arrive in the chart and you discuss them in the room. Most efficient, if check-in can support it.
In the visit, by interview. You ask the questions during the visit. Slower, more accurate for patients with literacy issues, language barriers, or cognitive concerns. You also hear how they interpret the question.
Avoid the third pattern: the patient fills it out and the clinician glances without engaging. Then the screener is a checkbox.
## How to score
Validated cutoffs:
PHQ-9:
* 0-4: minimal depression
* 5-9: mild
* 10-14: moderate
* 15-19: moderately severe
* 20-27: severe
GAD-7:
* 0-4: minimal anxiety
* 5-9: mild
* 10-14: moderate
* 15-21: severe
PHQ-9 item 9 is a suicide-risk question. Any positive response is a clinical event. Document the C-SSRS or your standard risk assessment, regardless of the total score.
The total is one input. The item pattern is another. Two patients with a PHQ-9 of 12 can look different: high scores on items 1-4 (mood, anhedonia, sleep, energy) vs high scores on items 7-9 (concentration, psychomotor, suicidality). Look at the items.
## How to chart
Keep the score as a structured field, not a sentence in the HPI.
Trending. A structured score can be charted over time in the record and in practice-level reporting. Free-text "PHQ-9 of 12" in the HPI cannot.
Reporting. Payer-mandated MBC reports query structured scores. They do not parse free text reliably.
In Nextvisit, PHQ-9 and GAD-7 (plus AUDIT, CRAFFT, PCL-5, and the practice-configurable list) are structured fields. The score lands when responses are captured. The trend graph shows in the chart header. Aggregate reporting rolls up across providers.
If your system does not support this, use a dedicated EHR-template section with score and date, separate from HPI prose. Keep the format consistent across visits.
## How to trend
A trend is more useful than any single score. Watch for:
* Sustained improvement. Two consecutive scores down 5 or more points typically reflects clinically significant improvement, often the "responder" threshold.
* Sustained worsening. Two consecutive scores up 5 or more points warrants a treatment-plan revisit.
* Plateau at a partial response. Stuck at a moderate score (10-14) for 6 to 8 weeks despite an adequate trial. Usually the signal to consider augmentation or rotation.
* Disconnects. The score does not match the clinical impression. The patient's function can improve while reported symptoms remain. Reported symptoms can improve while function remains impaired. Discuss the disconnect during the visit.
## What MBC reporting looks like
Aggregate reports of score trends can include:
* Number of patients screened in the quarter.
* Distribution of baseline scores.
* Percentage of patients with at least one follow-up score within 90 days of baseline.
* Distribution of change from baseline among patients with follow-up scores.
The reports want the data, not narrative. Structured scores come out of the warehouse. Free-text HPIs become a manual chart review.
## A note on patient experience
Some patients find the screeners mechanical. The mitigation is to talk about the score in the room, briefly. "Your PHQ-9 is 14 today, up from 9 last visit. That tracks with what you described about sleep and energy. Let's talk about whether we adjust the dose." The screener becomes part of the conversation. Adherence and accuracy both improve.
================================================================================
URL: https://nextvisit.ai/guides/privacy-42-cfr-part-2
Title: Privacy and 42 CFR Part 2, what changes in your charting
Description: What 42 CFR Part 2 requires for SUD records: who it covers, consent, segregation, disclosure logs, and how it changes the workspace.
Date: 2026-03-05
Audience: all
Reading time: 5 min
================================================================================
42 CFR Part 2 governs confidentiality of substance use disorder (SUD) treatment records. It predates HIPAA, applies on top of HIPAA, and in several ways is stricter. For practices that treat SUD as part of a broader panel, the rule shapes what goes in the chart, who can see it, and what has to happen before a record leaves the building.
## What Part 2 covers and what it does not
Part 2 covers records of patients receiving SUD diagnosis, treatment, or referral from a "Part 2 program." That definition is what trips practices up. It includes:
* Federally-assisted programs whose primary function is SUD treatment (an OTP, a residential SUD facility, a buprenorphine clinic).
* Identified units within general medical facilities where SUD diagnosis, treatment, or referral is the primary function.
* Medical personnel within general medical facilities whose primary function is SUD diagnosis, treatment, or referral.
A general outpatient psychiatry practice that treats some SUD patients alongside everything else is typically not a Part 2 program in the strict definition. A clinician in that practice whose primary function is SUD treatment may be subject to Part 2 for the records they create. Most MAT programs are. Most outpatient psychiatry practices that include MAT services are partially.
If you are not certain whether Part 2 applies, ask compliance counsel before you assume either way.
The 2024 update brought Part 2 closer to HIPAA in several places (notably a single broad consent for treatment, payment, and operations) but kept the core protections and added new ones around segregation, redisclosure, and breach notification.
## What stays the same in your charting
The substance of the SUD note does not change. Assessment, diagnosis, medication, monitoring plan, C-SSRS, toxicology: same documentation. Part 2 governs handling, not content.
Do not "thin" Part 2 records to reduce disclosure risk. A thin chart that fails medical necessity creates denial, audit, and malpractice exposure without solving the disclosure problem. Write the full clinical content. Handle it rigorously.
## What changes in your charting
Tag the record as Part 2. Most platforms use that tag to trigger segregation, audit, and disclosure controls.
Document the consents. Part 2 records can only be disclosed with patient consent in a specific format, with limited exceptions. The chart should make obvious which consents are in place, what they cover, and when they expire.
Mark the redisclosure prohibition. When a Part 2 record is disclosed, the material must carry a notice prohibiting redisclosure without further consent. Most platforms apply the standard wording on a Part 2 export.
Log every disclosure: date, recipient, records disclosed, consent used, and the redisclosure-prohibition notice that went with it. The log is part of the patient's record and available on request. A referral, a release to a PCP, and a carrier query are each a separate entry. Build the log into the release workflow. Practices that try to backfill it typically have an incomplete log.
## What changes in your workspace setup
Four operational patterns distinguish a Part 2-aware workspace from a HIPAA-only one:
* Segregation. Part 2 records are segregated so disclosures of non-Part 2 records do not include Part 2 content. That can be section-level or record-level.
* Consent management. Consent forms sit with the patient record. An external disclosure does not go through without an active, in-scope consent.
* Audit logging. Every access is logged: who, when, what they did.
* Workforce training. Annual training, documented per workforce member, on top of HIPAA.
Default to tagging SUD-related encounters as Part 2 unless the clinician affirmatively declasses the encounter for a documented reason. Tie consents to disclosures so a release cannot leave the system without a consent record attached.
In Nextvisit, the usual pattern is encounter-level tagging plus a custom tag (e.g., "Part 2") and export controls that exclude tagged content from default flows. Consents and disclosure logs live as documents at `/patient/[uuid]/documents` with structured fields for consent type, scope, and expiration.
If you are a dedicated SUD program (OTP, residential, embedded MAT clinic), talk to compliance counsel about a separate workspace or a more formal segregation architecture.
## The "co-occurring" problem
A patient with a primary mood or anxiety disorder and a substance use disorder, treated by the same clinician in the same encounter:
* Document the encounter in full, with the SUD content present and clinically complete.
* Tag the encounter. If the clinician is functioning as a Part 2 provider for this patient (primary function is SUD treatment, or SUD treatment is integral to the encounter's purpose), the encounter is Part 2. If not, the encounter is HIPAA-only and the SUD content still sits under HIPAA.
* Make the tagging basis visible. A short assessment note ("encounter tagged as Part 2 record per primary purpose of SUD treatment today") clarifies the call for a later reviewer.
* Tell the patient the Part 2 protections in plain language.
If the practice is not certain on a per-encounter basis, the safer default is to apply Part 2 more broadly.
## What changes in your AI documentation tooling
Part 2 records cannot be used to train AI models without specific patient consent in the Part 2 format. Nextvisit does not train AI on PHI, which includes Part 2 records. Nextvisit trains AI on other customer data by default, and customers can opt out. Confirm the same of any other tool and document the confirmation.
Longitudinal views (AI Timeline, Treatment Pulse, peer review) operate across the chart, including Part 2 content. Those views are access-controlled and audit-logged. If a clinician cannot read the underlying records, the derived view is also restricted.
AI Tasks that touch patient or encounter data may produce outputs with Part 2 content. The destination has to match the consents in place. A referral letter should not include Part 2 content unless the referral has Part 2-compliant consent.
MCP and OAuth are external disclosures. Part 2 records should not flow there unless the consent and the external party's compliance posture support it.
## Patient rights
Part 2 patients have the right to know how their records can be used, to consent to disclosures with clear scope and expiration, to revoke that consent, to access the disclosure log and their own records, and to a complaint process. Document acknowledgment at intake or at the first SUD-related encounter.
## Where Nextvisit fits
The platform supports the handling: encounter tagging, custom tags for Part 2 cohorts, document storage for consents and the disclosure log, audit logging on every access, segregated export controls, workspace-scoped API and MCP access. Clinical content is unchanged.
================================================================================
URL: https://nextvisit.ai/guides/onboarding-five-provider-group-week-one
Title: Onboarding a 5-provider group practice in week one
Description: A five-day plan for a small group: workspace setup, provider invites, templates, automations, and the week-one metrics that predict adoption.
Date: 2026-02-14
Audience: admin
Reading time: 4 min
================================================================================
At five providers, get everyone charting through the system by the end of day five, with the office manager as workspace owner and a handoff to a week-two rhythm. Times assume one administrator and five clinicians. Adjust as needed.
## Before day one
Two things should be in place before Monday morning.
The workspace exists and the administrator has owner access. Sign up at `nextvisit.app/register`. Confirm clinic profile, NPI, and tax ID under Settings > Clinic. Confirm at least one Service Location. If the practice has multiple locations, configure all of them. The Service Location Assigned event drives a lot of downstream automation.
The BAA is signed. Have the provider list, credentialing data, and any payer enrollment ready to import. If the practice uses the insurance billing module, credentialing and payer-network data live under `/apps/insurance`.
Block one hour on each provider's calendar on day one. Block 30 minutes on days two through five. Five hours of provider time across the week is the rough target. Block two hours on the administrator's calendar each day.
## Day 1 (Monday): workspace foundation and provider invites
Morning is administrator time.
* Confirm clinic profile, providers, and Service Locations under Settings > Clinic.
* Generate any required API keys under Settings > Tools and AI > API Keys. Most groups do not need API access in week one; document the path for later.
* Decide naming conventions for templates, automations, and patient tags. Write them down. Consistency keeps the workspace from sprawling.
* Set up the provider list and send invitations.
Afternoon is provider time. Each provider gets a 60-minute session:
* 10 minutes: workspace tour and credit-meter awareness.
* 20 minutes: New Recording flow. The provider records a sample visit (real or simulated) end to end and watches the note generate.
* 20 minutes: editing the draft, signing the encounter, watching the claim auto-create on the insurance side if applicable.
* 10 minutes: questions and the next-day plan.
End-of-day check: every provider has signed at least one encounter, understands the mobile flow at a basic level, and knows where the credit meter is and roughly what daily usage will look like.
## Day 2 (Tuesday): chart in the system, tune the defaults
First full clinical day. Providers chart their normal panel through Nextvisit on the default templates.
The administrator observes. Walk the floor (or join a mid-day video call) and ask three questions per provider:
* Are the drafts close to what you would have written, or far off?
* What are you editing on every chart?
* Is credit usage per visit roughly what you expected?
Capture the answers. Recurring edits are the day-three target. Far-off drafts on a visit type are the day-four target.
End-of-day check: every provider has charted a normal half-day or full day. The administrator has a list of recurring per-provider edits.
## Day 3 (Wednesday): structural settings and personalization
Each provider spends 30 minutes on the structural levers.
* Style preferences in Settings > Account: bullets vs prose for MSE, "no SI reported" vs "denies SI," default follow-up format.
* Coding behavior under Settings > Tools and AI > Insurance Coding: confidence threshold, default code sets, specialty-specific rules.
* If a recurring edit is structural rather than stylistic, capture it as a setting now. Five minutes removes the edit from every later chart.
The administrator reviews the credit meter, the encounter list, and claims that auto-generated on day two. Confirm that signed encounters with ICD-10 codes flowed into `/apps/insurance/claims`. If they did not, fix it today, before the week's billing depends on it.
End-of-day check: providers report a visible drop in editing time vs day two. Administrator has confirmed the encounter-to-claim flow.
## Day 4 (Thursday): custom templates for the visit types that need them
Most providers have one or two visit types the defaults handle poorly. Spend 30 minutes per provider in the template editor at `/apps/custom-prompts`.
* The provider brings a real recent example of the visit type, anonymized.
* The provider and administrator (or the provider alone) build the template using the "Build a custom intake template" guide.
* The provider applies the template to the next two visits of that type and edits the drafts.
* By end of day, each provider has at least one custom template running.
If the practice has cross-provider visit types (Spravato, TMS, MAT, lithium monitoring), build those once and share them. Shared templates beat per-provider ones.
End-of-day check: at least one custom template per provider, plus any practice-wide templates, are saved and in use.
## Day 5 (Friday): automations, review, and the operating rhythm
Set up the automations that save the most admin time. Then review the week and lock the operating rhythm.
Three automations worth setting up in week one:
* A scheduled task that aggregates unsigned encounters older than three days, with provider names, due to the administrator. Monday at 8 AM. Intelligence level Fast. Patient/encounter tools on.
* An event-based task on Encounter Signed that drafts a referral or follow-up letter when the chart contains specific markers (a new diagnosis, a referral order). Specialty-dependent.
* An event-based task on Document OCR Complete that summarizes inbound records into a one-paragraph chart insert. The administrator reviews and routes.
Tasks live at `/apps/tasks`. Default to Normal intelligence. Reserve Ultra for complex weekly aggregates and Budget for high-volume, low-stakes routing.
Fifteen minutes with each provider:
* Editing time per chart, day one vs day five.
* Any visit type still producing far-off drafts.
* Settings or templates that need a second-round adjustment.
* Confidence heading into week two.
The administrator's review: signed encounters vs last week, anything stuck in Processing or Ready more than 48 hours, claims that did not auto-generate, credit usage, and provider feedback.
End-of-day check: routine visits are signing with little editing, custom templates are saved, the first three automations are running, and week-two priorities are written down.
## What week-one success looks like
By Friday you should see:
* Editing time on routine visits down 60 to 80 percent from Monday.
* Every provider charted through the system on every clinical day.
* Custom templates running for the visit types defaults missed.
* Encounter-to-claim running without workarounds.
If three of five providers hit those marks, week one succeeded. If two or fewer did, week two is coaching, not scaling.
## Week two and beyond
Daily: chart every clinical visit. Review unsigned encounters and stuck claims. Weekly: credit meter, claims pipeline, provider feedback. Monthly: hours saved, same-day claims, denials, and any payer-mandated MBC scores.
================================================================================
URL: https://nextvisit.ai/guides/building-patient-timelines-for-better-psychiatric-care
Title: What a psychiatric timeline should hold
Description: What a psychiatric timeline should hold (scales, meds, diagnoses, session context) and how to add one without replacing the EHR.
Date: 2025-12-23
Audience: psychiatrist, np-pa, therapist, admin
Reading time: 2 min read
================================================================================
## 
A psychiatric timeline is a visual summary of symptoms, medications, diagnoses, and session context across visits. Instead of reading notes one at a time, you see months of care in one view.
Psychiatric patients often have dozens of visits a year. Visit-by-visit notes hide the trend.
## What to put on it
- PHQ-9 and GAD-7 for symptom trajectory. Add C-SSRS, MDQ, or ASRS when the diagnosis calls for it.
- Medication history: start and stop dates, dose changes, documented side effects
- Diagnoses and rule-outs with dates of establishment, remission, or revision
- Session context: stressors, safety, therapy focus, functional change
Together those can show activation after an SSRI titration, or sleep improvement after CBT-I, that isolated notes hide.
ONC's [United States Core Data for Interoperability (USCDI)](https://www.healthit.gov/isa/united-states-core-data-interoperability-uscdi) now includes [mental and cognitive functioning elements](https://www.healthit.gov/isa/uscdi-data-class/mental-cognitive-functioning). That is the federal direction for behavioral health data.
## How to add one without replacing the EHR
1. Generate the timeline from existing notes and structured fields.
2. Keep a lightweight data model: date, event type, value, and note link.
3. Copy or import summaries back into the chart as needed.
Clinics with deeper integration can use FHIR reads for medications, problems, and observations. Limit access by role, log views and edits, and keep exports inside HIPAA policy. HHS covers the [technical safeguards for systems handling ePHI](https://www.hhs.gov/hipaa/for-professionals/security/laws-regulations/index.html).
Pilot on the cases where the benefit is obvious: treatment-resistant depression, bipolar spectrum, and ADHD or anxiety comorbidity. Track time to insight before a medication change, how often prior notes get opened during pre-chart, and after-hours charting time.
================================================================================
URL: https://nextvisit.ai/guides/psych-soap-notes-that-prove-medical-necessity
Title: Psychiatry SOAP notes that support medical necessity
Description: Medical necessity is the right signal, not a long note. How to chart S, O, A, and P so a reviewer can follow the visit.
Date: 2025-12-23
Audience: psychiatrist, np-pa, therapist
Reading time: 4 min read
================================================================================
In psychiatry, a SOAP note has to connect today's complaints and observations to an assessment that justifies the plan. Another clinician should be able to read it. An auditor should be able to defend it. You should be able to reuse it next visit.
The APA's record-keeping guidance emphasizes clarity, continuity, and appropriate detail. See [APA Record-Keeping Guidelines](https://www.apa.org/practice/guidelines/record-keeping). Aim for sufficiency, not length.
## Subjective
Capture the patient's words and the interval: change since last visit, adherence, side effects, function at work, school, or home, and safety. Skip filler like "patient doing ok." Be specific enough to compare visits.
## Objective
Include the mental status exam (appearance, behavior, speech, mood/affect, thought process/content, cognition, insight/judgment), vitals if taken, and standardized scales when used.
If you use PHQ-9 or GAD-7, include:
- Total score
- Severity band
- Item 9 status
- Change from prior
NIMH's medication overview is a useful companion for patient education: [NIMH: Mental Health Medications](https://www.nimh.nih.gov/health/topics/mental-health-medications).
## Assessment
This is medical decision-making. State diagnoses (new or established), progress toward goals, differential, risk, and how the data informed the conclusion. Tie symptoms and scales to function. Document why you are continuing, changing, or stopping treatment.
## Plan
Spell out medications or therapy, labs or monitoring, education given, follow-up interval, and contingencies. Example: "If activation persists beyond 72 hours, message clinic."
For privacy expectations, see [HHS Summary of the HIPAA Security Rule](https://www.hhs.gov/hipaa/for-professionals/security/laws-regulations/index.html).
---
# Psychiatry SOAP note examples for med management
Condensed examples you can adapt.
## Example 1: Depression follow-up with partial response
**Subjective:** "Energy is better; still waking 3 to 4 a.m." Works full-time. Missed two doses this week. No SI. PHQ-9 = 11 (was 15), item 9 = 0. Side effects: mild nausea first week, now resolved.
**Objective:** MSE: cooperative, tearful at times; speech normal; mood "tired"; affect constricted; TP linear; TC no SI/HI/AVH; I/J fair. Vitals WNL.
**Assessment:** MDD, recurrent, moderate. Partial response to SSRI with residual insomnia and anergia. No acute safety concerns.
**Plan:** Increase SSRI 20 to 30 mg daily. Sleep hygiene review. Consider CBT-I resources. Follow-up 3 to 4 weeks. Contingency: if activating, reduce to 20 mg and message. Education on black-box warning provided. See [NIMH: Mental Health Medications](https://www.nimh.nih.gov/health/topics/mental-health-medications).
## Example 2: Generalized anxiety with therapy focus
**Subjective:** "Worry spirals most evenings." GAD-7 = 14 (was 13). No panic attacks. Using skills inconsistently.
**Objective:** MSE: anxious but engaged; no psychosis; cognition intact. Last PHQ-9 = 8.
**Assessment:** GAD, persistent. Psychotherapy primary. Consider SSRI augmentation if no improvement.
**Plan:** Intensify CBT with exposure hierarchy. Skills practice 10 min nightly. No med changes. Follow-up 2 weeks. Provide handouts.
## Example 3: ADHD med check with appetite concerns
**Subjective:** Focus improved. Appetite lower at lunch. Weight stable. Sleeps 7 hours. No palpitations.
**Objective:** BP/HR WNL. MSE unremarkable. Work performance improved per supervisor email.
**Assessment:** ADHD, combined type. Good response. Mild appetite suppression.
**Plan:** Continue dose. Add protein breakfast. Monitor weight. Follow-up 1 month. Consider long-acting switch if appetite worsens.
Copy the skeleton. Change the specifics. Necessity shows up as trajectory, function, risk, and a plan with monitoring.
---
# How to turn these examples into faster workflows
## Make the structure predictable
Run the same compact SOAP every visit:
- Subjective: interval change, adherence, side effects, function, safety
- Objective: MSE with free-text nuance for anything abnormal
- Assessment: diagnoses, progress, differential, and risk
- Plan: medications (dose, start date, monitoring), therapy, education, follow-up, contingencies
## Embed SOAP structure where it helps
- Surface the last 3 to 6 months of PHQ-9/GAD-7 trends next to the note composer.
- Pre-fill med names with generic/brand, dose, and last change date.
- Auto-insert scale interpretation text with severity bands and change from prior.
## Measure outcomes
Track time to sign, denials citing documentation, addenda per note, and after-hours minutes.
================================================================================
URL: https://nextvisit.ai/guides/top-features-to-look-for-in-an-ai-scribe-for-psychiatry
Title: What a psychiatry AI scribe has to capture
Description: What a psychiatry AI scribe has to capture: DSM-5 templates, live notes, HIPAA handling, EHR export, and audit-ready outcomes.
Date: 2025-12-23
Audience: all
Reading time: 2 min read
================================================================================
## DSM-5 and specialty templates
A general-purpose AI scribe will not reliably capture psychiatric documentation. Templates need to hold symptoms, the mental status exam, diagnostic impressions, and the plan in the right places.
Use structured formats for medication management, psychotherapy, and integrated behavioral health so those fields stay consistent visit to visit.
## Live notes and structured fields
The scribe should transcribe and structure history, symptoms, progress, and interventions during the visit so after-hours charting shrinks.
Structured fields also make analytics, outcome tracking, and billing more reliable than a wall of prose.
## HIPAA and patient privacy
HIPAA is the floor. Look for encryption, auditable access controls, and a clear path for consent, recording, note generation, and export.
## EHR export that does not add a second chart
The scribe should export or sync structured notes, medication lists, and diagnostic codes into the EHR you already use. Adoption drops when providers have to change how they chart.
## Outcomes and an audit trail
You need longitudinal timelines, outcome measures, and an audit trail on every encounter. That is what chart review and payer documentation actually ask for.
================================================================================
URL: https://nextvisit.ai/guides/ai-compliant-upcoding-mental-health-revenue
Title: Reviewing mental health service codes
Description: Review E/M codes 99213-99215, add-on codes 90833, 90836, and 90838, and standalone codes 90834 and 90837.
Date: 2025-11-19
Audience: admin, psychiatrist, np-pa
Reading time: 4 min read
================================================================================
Review each code against the documented service and current payer rules. Coding tools can flag possible mismatches for clinician review.
## Match the code to documented work
Select codes 99213 through 99215 by documented medical decision-making or total time when billed without a psychotherapy add-on. When you bill 90833, 90836, or 90838, select the E/M level by medical decision-making. Do not include psychotherapy time in the E/M selection.
## Match the billed level to the documentation
The record must support the selected E/M level. Review each suggestion before you apply it.
## Psychotherapy add-on codes
Codes 90833, 90836, and 90838 apply to psychotherapy provided with an E/M service. The psychotherapy must be significant and separately identifiable.
Document the psychotherapy time. Code 90833 covers 16-37 minutes, 90836 covers 38-52 minutes, and 90838 covers 53 minutes or more.
Review [CMS office and outpatient E/M guidance](https://www.cms.gov/sites/default/files/2022-01/Physician_Fee_Schedule__PFS__Payment_for_OfficeOutpatient_Evaluation_and_Management__EM__Visits.pdf), [CMS psychotherapy guidance](https://www.cms.gov/medicare-coverage-database/view/article.aspx?articleStatus=A&articleid=57480&bc=5&contractorNumber=all&sortBy=title&stateRegion=s14&ver=31), and current payer rules before billing.
## E&M code selection: 99213 vs 99214 vs 99215
For codes 99213 through 99215, review the documented medical decision-making or total time. If you bill a psychotherapy add-on, use medical decision-making to select the E/M level.
## Psychotherapy code review
Code 90834 covers 38-52 minutes. Code 90837 covers 53 minutes or more. Document the session time and the psychotherapy service.
Review intervention language, treatment plan progress, risk assessments, and time before you select a code.
## Documentation requirements
The note has to support every code billed.
For an E/M level, document the medical decision-making or total time used to select the code. If you bill a psychotherapy add-on, document the separate service, intervention, response, and psychotherapy time.
Update the treatment plan when it changes. Complete a risk assessment when clinically indicated. Give a rationale for each medication change.
## Review each visit for coding accuracy
Compare each billed code with the documentation. Review each mismatch. Apply a code only when the documentation and payer rules support it.
================================================================================
URL: https://nextvisit.ai/guides/choosing-the-best-ai-scribe-for-psychiatry-practices
Title: What matters in a psychiatry AI scribe
Description: Evaluate psychiatry AI scribes for specialty templates, risk language, and session-to-session context.
Date: 2025-10-14
Audience: all
Reading time: 3 min read
================================================================================
Evaluate how each AI scribe handles psychiatric documentation. Check its templates, risk language, and context from earlier sessions.
Typical gaps:
- DSM-5 structured assessments. Generic scribes do not organize symptoms into diagnostic criteria.
- Differential diagnosis tracking. Mental health diagnoses evolve and need a longitudinal view.
- Session-to-session context. A statement in session 12 only makes sense with sessions 1-11.
- Risk assessment language. Suicide and homicide risk get flattened.
- Verify that the product separates psychotherapy notes from progress notes and supports the required handling for each type.
That gap is liability, not inconvenience.
## What to ask for
### Specialty templates, beyond SOAP
Check support for SOAP, psychiatric intake, medication management, therapy progress, and crisis evaluations.
Ask vendors: "Show me an actual note from a patient with comorbid depression and anxiety who is on three medications." If it does not look like something you would write, keep looking.
### HIPAA that is more than a badge
Ask about:
- A signed BAA. If they will not sign one, walk away.
- Encryption at rest and in transit.
- Who at the vendor can see patient data. The answer should be almost nobody, and they are audited.
- What happens to recordings after the note is generated. They should be deleted.
- How they notify you of a breach.
Read the security documentation.
### Clinical intelligence beyond transcription
Transcription is the floor. Psychiatry needs medication follow-up across visits, PHQ-9 trajectory, which interventions actually worked, and a differential that does not reset every session.
## Cost vs time
Compare current prices, contract terms, and trial results with your practice requirements. During the trial, measure draft quality, review time, workflow fit, and support.
## Mistakes that waste the first month
- No review period. Plan 2-3 weeks of checking every note.
- Leaving the default templates alone. Customize them to how you actually practice.
- Trusting it too early. Your license is on the line.
- No consent policy. Some patients do not want recording. Have an alternative so the tool does not get in the way of the alliance.
## Questions before you buy
1. Can I see actual psychiatric notes, not marketing samples?
2. How long before providers see time savings?
3. Show me notes from patients with multiple comorbidities.
4. What is the cancellation policy?
5. Who uses this: solo, groups, or both?
If you want to see how Nextvisit handles this, [book a demo](https://nextvisit.ai/demo) or [start a free trial](https://nextvisit.app/register) and test it on your own panel.
# Help center
================================================================================
URL: https://nextvisit.ai/help/uploading-documents
Title: Uploading Documents to Patient Charts
Description: Upload PDFs to a patient's chart and attach them during recording or note writing.
Date: 2026-09-06
Topic: Patient charts
================================================================================
Upload PDFs from a patient's chart or during a recording session. **In the recorder, select a patient before uploading.**
## Upload options
**From the patient chart:**
Open the patient's chart, go to **Documents**, and click **Upload**. Drop a PDF or click to browse.
**During a recording session:**
Link your patient first, then open **Files** (or press **F**) and drop a PDF. You can upload during the session.
## Using previously uploaded documents
When writing a note, uploading audio, or starting a new recording session, you can select documents already in the
patient's chart. The patient must be linked to access these files.
Useful uploads: lab work, intake forms, prior records, imaging reports, specialist consultations.
## File requirements and credit usage
Uploads must be PDF files. Document processing uses credits. Credit terms depend on your clinic's agreement.
Remaining credits are on [Usage and credits](/help/usage).
================================================================================
URL: https://nextvisit.ai/help/recording-a-session
Title: How to record a live session
Description: Record a live session: pick a template and patient, transcribe, add scales and codes, then End. Includes Telehealth Mode.
Date: 2026-09-04
Topic: Notes and AI
Reading time: 8 min read
================================================================================
import MockFigure from '~/components/patterns/MockFigure.astro';
import LiveTranscriptionMock from '~/components/patterns/recorder/LiveTranscriptionMock.vue';
import TemplateSelectMock from '~/components/patterns/recorder/TemplateSelectMockBound.astro';
import SessionSetupMock from '~/components/patterns/recorder/SessionSetupMock.vue';
import SidePanelMock from '~/components/patterns/recorder/SidePanelMock.vue';
import AnalyzingPipelineMock from '~/components/patterns/recorder/AnalyzingPipelineMock.vue';
## Before you start
The recorder runs in any modern browser at `nextvisit.app/recorder`. The first time you start a session, the browser asks for microphone access. Approve it once. Telehealth visits also prompt you to share the call's browser tab when you click Start Recording. That share prompt is per session and comes back on every pause-and-resume cycle. Nothing to install.
## Step 1: Open the recorder and pick a template
Open the recorder and the template gallery loads. Your most recent pick is pinned at the top under **Recently Used** for one-click reuse. Below that, **Your Templates** shows any custom templates your workspace has authored. The full library lives in the **All Templates** grid, with a search input that filters live as you type.
Click a card to advance to session setup. If you grabbed the wrong template, the chosen one appears as a chip at the top of the next screen, and clicking it sends you back to swap.
## Step 2: Pick a patient and review the auto-populated context
Search for the patient by name in the picker at the top of session setup. Selecting a record auto-populates the pre-session context editor below it: today's date, the logged-in provider, the patient's name and DOB, and an active medications table pulled from the chart. The editor is a full rich-text surface, so you can edit the populated fields, add a sentence about why the patient is in today, or paste in anything else AriaMD should ground the note on.
If you want the AI to reference specific patient documents, expand **Link Documents** and pick up to 5. They feed into the AI's context silently during recording and appear as a **Documents** tab on the encounter detail when the session ends.
The **Capture** card on this screen has **Telehealth mode**. The line under the label reads: Capture both sides of a virtual visit through your screen's audio. Turn the switch on at session setup, before **Start Recording**.
## Step 3: Start recording
Click **Start Recording**. The view switches to live transcription, the status dot turns red, the timer starts, and your first words begin to appear. As you speak, the in-progress phrase builds in cyan below the latest finalized line, with a blinking cursor. Finalized lines stamp themselves with the wall-clock time and stack above.
## Add context, scales, files, and codes mid-session
The right-hand action panel stays available throughout the session. Six tools, each one click away, without leaving the recorder.
**Add Note** drops a non-spoken context note into the transcript at the current timestamp. Use it for things the AI should know but should not transcribe verbatim, like "patient nodded but did not verbalize agreement" or "appears alert and oriented x3, no acute distress." Submitted notes pin to the top of the transcript as a labeled CONTEXT NOTE block, separate from the live conversation.
**Scales** opens the measure picker. Pick an instrument (PHQ-9, GAD-7, AIMS, MMSE, COWS, and the rest), add a sentence of context in the Notes field, and click **Generate with Context**. The AI scores the instrument from the in-progress transcript plus your notes and returns a filled preview with the total, the per-item responses, and the severity interpretation. Confirm to insert the completed scale into the transcript. After the encounter is signed, the measure also appears in a dedicated Scales & Exams panel on the encounter detail.
**Coding** lets you pre-stage ICD-10 diagnostic codes and billing codes during the visit, before AriaMD's auto-suggestions land on the encounter detail. Type a code or a description, multi-select from the dropdown, and save. Selected codes appear as removable pills above the search.
The remaining three tools live in the same panel:
- **Context** opens the same rich-text editor you used at session setup, so you can edit the AI's grounding mid-session. Saves are live.
- **Files** uploads a PDF (lab results, genetic tests, or other documents) for AriaMD to reference. Link a patient first. This is separate from the Link Documents picker on session setup, which sources files already in the patient's chart.
- **Audio** opens audio settings to change microphone or toggle echo cancellation, noise suppression, and auto gain control mid-session.
## Pause and resume
Press `Space` or click the Pause button to pause. The status dot flips to yellow, the listening pill changes to **Paused**, and the transcript stays put. Press `Space` again or click **Resume** to pick back up. Audio settings changes only take effect after a pause-and-resume cycle, which is the supported way to switch microphones mid-visit.
## Keyboard shortcuts
| Key | Action |
|---------|---------------------------------|
| `Space` | Pause / Resume |
| `N` | Add a context note |
| `F` | Open Files (Upload Attachments) |
| `⌘F` | Search transcript |
| `⌘.` | End the recording |
## When you are done
Click **End** or press `⌘.` to finish. The recorder hands the session off to the encounter detail page and runs the AI analysis pipeline in front of you: Creating Timeline Events, Generating Note, Suggesting Codes, and so on, up to seven steps when documents are linked. The card turns green at the end with **Encounter Processed**.
The encounter detail page renders with **Note**, **Transcript**, **Peer Review**, and **Context** tabs, plus a **Documents** tab when files were linked at setup. A coding sidebar shows AriaMD's ICD-10 and billing suggestions alongside anything you pre-staged. Scales and exams from the session appear in their own sidebar panel. For more on what AriaMD does to your session after you click End, see the [features overview](/features#aria).
On an unsigned note, the row **Tell Aria what to change…** edits that draft. See [Edit a note with Aria](/help/edit-a-note-with-aria).
## Telehealth mode
On Zoom, Doxy, Teams, or Google Meet, a mic-only recording catches the other side as muffled speaker bleed. Telehealth Mode captures the patient's audio from the call's browser tab at line level and mixes it with your microphone.
### Turning it on
At session setup, turn on **Telehealth mode** in the **Capture** card. With the switch on, this notice appears: When recording starts, pick "Share system audio" or "Share tab audio" in the browser prompt so both sides of the visit are captured.
When you click **Start Recording**, the browser's native screen-share picker appears. Pick the tab running the video call and check **Share tab audio** (Chrome's wording; some browsers say **Share system audio**). Confirm. Recording begins, and the transcript captures both sides of the call.
If you forget to check the audio box, the recorder catches the empty audio stream and silently falls back to microphone-only recording. The session continues, but the line-level patient capture is gone for the rest of the visit.
### Pause and resume re-prompts the share dialog
Each pause-and-resume cycle tears down the screen capture and re-runs the picker. On long telehealth visits with several pauses, the share dialog comes back every time you resume. Pick the same tab and check the audio box again to keep capturing both sides.
### If you click "Stop sharing" mid-session
The browser shows a persistent "Stop sharing" indicator while you are sharing the tab. Clicking it does not stop the recording, but the patient's audio drops out from that point and the recorder continues with microphone-only audio for the rest of the session. To restore patient capture, end the recording and start a new one.
### Browser support
Telehealth Mode requires desktop Chromium browsers: Chrome, Edge, Brave, Arc, Opera. Firefox shows the toggle but the share picker does not offer an audio checkbox, so the recorder falls back to microphone-only. Safari support is partial and inconsistent. The toggle is not rendered on mobile (iOS Safari and Android Chrome do not support tab-audio capture).
### Privacy
In real-time recording mode, no audio file is saved on either Nextvisit's servers or the video platform's. Only the text transcript is persisted. The video call platform itself (Zoom, Doxy, Teams) is unaffected: Telehealth Mode is browser-level audio capture from the tab you point it at, not an integration with the video service.
## Troubleshooting
### Mic not detected
Check the browser's site permissions for `nextvisit.app` and confirm microphone access is granted. If it is, open **Audio Settings** from the gear icon, switch to a different microphone in the dropdown, and click **Test Mic** to confirm levels move. If the recorder is already running and you change microphones, pause and resume to apply the change.
### Transcript stops mid-session
Check the listening pill in the upper right. If it flipped to **Paused**, press `Space` to resume. If it shows a connectivity error, the recorder queues audio locally and resumes transcription automatically when the connection returns. Nothing is lost during the gap.
### Recording ended unexpectedly
All sessions auto-save up to the moment of disconnect. Open the recorder again, pick the same patient, and the in-progress encounter appears under **Resume**. Pick up where you left off, and AriaMD stitches the segments together when you click End.
================================================================================
URL: https://nextvisit.ai/help/api-keys
Title: Managing API Keys for Third-Party Integrations
Description: Create, view, and manage API keys for connecting external applications to your Nextvisit AI account.
Date: 2026-09-01
Topic: EHR and integrations
================================================================================
## Accessing API Keys
1. Click your **user profile** in the bottom left corner
2. Click **Settings** (gear icon)
3. Select **API Keys** from the menu
The page heading is **API Tokens**. **Create Token** adds a token. **Manage Tokens** appears after you have at least
one.
OAuth apps (Claude, MCP, and other authorized clients) are under **External Apps** in Settings, not on this page.
## Creating and managing tokens
Create a token only for an application you trust. Each token grants access to your Nextvisit account data. Open
**Manage Tokens** to see token names and revoke a token you no longer need.
## Rate limits
API requests are limited to 240 per minute. Contact support if your use case requires a higher limit.
================================================================================
URL: https://nextvisit.ai/help/mcp-server
Title: Nextvisit MCP Server
Description: Connect Claude, ChatGPT, or a local LLM to Nextvisit over MCP. Endpoint, OAuth, tools, scopes, and troubleshooting.
Date: 2026-09-01
Topic: EHR and integrations
================================================================================
MCP-aware assistants connect once, then read structured chart data and write back through typed tools. The same
workspace scope as API keys and OAuth applies. PHI never leaves the workspace boundary.
## Server endpoint
```
https://nextvisit.app/mcp/clinic
```
Transport is HTTPS. Authentication is OAuth 2.0: your MCP client opens a browser, you sign in to Nextvisit and approve
the connection in the same consent screen used for OAuth applications, and the client stores the resulting token. There
is no API key to paste, copy, or rotate by hand. The server supports the three MCP operation classes:
- **Interactive**: tools that return an embedded Nextvisit UI block (a patient card, a recent-patients list) inline in
the assistant's reply.
- **Read**: typed tools that fetch patients, encounters, notes, documents, medications, and timelines as structured
JSON.
- **Write**: typed tools that create patients, edit unsigned encounter content and timeline, link or unlink patient
associations, manage note instructions, and sign or unsign an encounter.
## What you need first
1. A Nextvisit account with access to the workspace you want the assistant to read or write.
2. An MCP-aware client. Claude desktop, Claude on the web, ChatGPT, Cursor, Continue, and most local LLM runtimes
(Ollama, LM Studio, Open WebUI) support remote MCP servers with OAuth.
If your workspace admin has disabled external app access for the workspace, the OAuth consent screen will refuse the
connection. Contact them or our support team to enable it.
## Connecting Claude
Claude's desktop and web clients both support remote MCP servers as **Custom Connectors**.
1. Open **Settings > Connectors** in Claude.
2. Click **Add custom connector**.
3. Name it `Nextvisit`.
4. Set the **MCP server URL** to `https://nextvisit.app/mcp/clinic`.
5. Click **Connect**. A browser window opens to the Nextvisit OAuth consent screen.
6. Sign in to Nextvisit (if you are not already signed in), pick the workspace you want the connector to see, and review
the scopes the assistant is asking for. Click **Allow**.
7. The browser hands you back to Claude. The tools surface under the Nextvisit connector and become available to any
Claude conversation where you enable the connector.
The same steps work anywhere Claude exposes a Custom Connector field. Revoke from **Settings > External Apps** in
Nextvisit. The page heading is **Authentication**.
## Connecting other clients
Any client that supports OAuth-authenticated remote MCP servers works. Enter the URL; the client drives the browser
flow.
- **ChatGPT**: Add as a custom MCP connector under **Settings > Connectors** in ChatGPT. Paste the URL and complete
the OAuth flow in the popup window.
- **Cursor, Continue, Zed, and other IDE clients**: Add the URL to the `mcpServers` block of the client's configuration
file. On first call the client opens the OAuth flow in your default browser and caches the resulting token.
- **Local LLM runtimes**: Ollama, LM Studio, and Open WebUI accept remote MCP endpoints in their MCP settings panel.
Each client supports OAuth-authenticated remote MCP servers; check the runtime's MCP docs for the current shape.
If your client only supports MCP servers that authenticate with a static bearer token, it cannot connect to the
Nextvisit MCP server. The endpoint is OAuth-only and does not accept long-lived API keys.
## Tools available
The MCP server ships the same tool surface that powers Nextvisit's own AI features. Grouped by operation class:
### Interactive
- **recent_patients**: Returns the most recently updated patients for the current workspace as an embedded UI block.
- **fetch_patient**: Look up a patient by MRN or UUID. Returns chart details and an embedded patient display.
### Read
- **fetch_encounter**: Detailed encounter information. View modes: `full`, `content`, `transcript`, `context`.
- **search_encounters**: Search by patient name, encounter content, or keyword. Supports date range filters and
pagination.
- **utility_global_search**: Full-system search across patients, encounters, and documents. Supports phrase search
(`[exact phrase]`) and negative search (`-term`).
- **fetch_patient_medications**: Current and historical medication list, including dosage, frequency, classification,
prescriber, and status.
- **fetch_patient_timeline**: Patient visit history as a chronological encounter timeline.
- **fetch_document**: Retrieve an uploaded document by ID. Supports full read with pagination, or in-document phrase
search.
- **manage_note_instructions** (action `get`): Read your custom AI note generation instructions.
- **manage_patient_chart_notes** (action `get`): Read a patient's persistent chart notes.
### Write
- **create_patient**: Create a new patient record. Always search first to avoid duplicates. First and last name are
required; demographics, allergies, medications, and persistent notes are optional.
- **edit_encounter**: Edit `content` (clinical documentation) or `transcript` on an unsigned encounter. Three
operations: `str_replace`, `insert` (before or after an anchor), and `regex`.
- **edit_encounter_timeline**: Edit or append events on an encounter timeline.
- **link_patient_to_encounter**: Link or unlink the patient association on an encounter.
- **sign_encounter**: Sign (lock) or unsign (unlock) an encounter. Only the encounter creator can perform this action.
- **manage_note_instructions** (action `update`): Replace, append, prepend, or clear your AI note instructions.
- **manage_patient_chart_notes** (action `update`): Replace, append, prepend, or clear a patient's persistent chart
notes.
The server returns structured JSON.
## What an assistant call looks like
A typical read flow from Claude:
```
User: "What did Dr. Patel see Jane Doe for in April?"
Claude calls search_encounters(query="Jane Doe", date_from="2026-04-01", date_to="2026-04-30")
→ returns a list of encounter IDs with dates and titles.
Claude calls fetch_encounter(id="enc_a1b2", view="content")
→ returns the merged clinical note for that encounter.
Claude summarizes the visit back to you.
```
A typical write flow:
```
User: "Add 'patient declined PHQ-9 today' to the timeline on encounter 41822."
Claude calls edit_encounter_timeline(encounter_id=41822, action="add",
find="Patient declined PHQ-9 today")
→ event appears on the encounter timeline.
```
Every call carries the workspace scope. The assistant cannot see or modify charts in another workspace.
## Permissions and audit
- Tool calls obey the same workspace boundary as the rest of the platform. No cross-workspace reads, ever.
- Scopes are granted at OAuth consent time and you can review them before approving. Read scopes unlock the read and
interactive tools; write scopes additionally unlock `create_patient`, `edit_encounter`, `edit_encounter_timeline`,
`link_patient_to_encounter`, `sign_encounter`, and the `manage_*` update operations. A tool call that exceeds the
granted scope returns a 403 and never modifies data.
- `sign_encounter` is further restricted to the encounter's creator.
- `edit_encounter` only works on unsigned encounters. Unsign first, edit, then re-sign.
- A signed encounter also blocks writes to medications, diagnoses, and billing codes until you unsign it.
- Every MCP call is recorded in the workspace audit log with the timestamp, the connected app, the tool name, and the
affected resource ID.
- You can revoke the connection at any time from **Settings > External Apps**. Revocation invalidates
the token immediately; subsequent calls from that client return 401 until you re-authorize.
## Privacy
The MCP server is part of your Nextvisit workspace and operates under the same Business Associate Agreement. PHI
returned to the assistant is governed by the assistant vendor's own data handling, so when you connect Claude, ChatGPT,
or any third-party model, your workspace's chart data is subject to that vendor's terms during the conversation.
For 42 CFR Part 2 workspaces, MCP access follows the same consent gates as OAuth and external apps. See the [42 CFR
Part 2 guide](/guides/privacy-42-cfr-part-2) for the full policy.
## Troubleshooting
### Authentication failed or 401 errors
The stored token expired or the connection was revoked. Disconnect and reconnect the Nextvisit connector in your client
to run the OAuth flow again. You can confirm the connection is still active under **Settings > External Apps** in
Nextvisit.
### OAuth popup blocked
Some browsers block the OAuth popup the first time. Allow popups for the client's domain (for Claude on the web, that is
`claude.ai`) and retry the connection.
### Tool list does not appear
The client cached an empty tool list. Disconnect and reconnect the Nextvisit connector, or restart the client. Custom
Connectors in Claude refresh on reconnect.
### "Cannot edit a signed encounter. Unsign it first."
Call `sign_encounter` with `status: "unsigned"` first (you must be the creator), make the
edits, then re-sign.
### "Patient not found"
Confirm the MRN or UUID is from the same workspace you approved when you connected the assistant. MRNs are workspace-scoped, so the same number can
exist in two different workspaces and point to different patients.
Contact support for setup, scopes, or a sandbox workspace if you do not want to test against live charts.
================================================================================
URL: https://nextvisit.ai/help/nextvisit-go
Title: What is Nextvisit Go?
Description: Nextvisit Go is a separate note workspace for non-prescribing therapists and counselors. How it differs from the main Nextvisit platform.
Date: 2026-09-01
Topic: Getting started
================================================================================
Nextvisit Go is a separate product for non-prescribing providers, including therapists and counselors. Record a session, watch the transcript stream, generate a SOAP, DAP, or BIRP note, edit it with Note Copilot, and copy it into the EHR you already use.
Nextvisit Go does not include prescribing or medication-management workflows. Prescribers use Provider+ or Enterprise on the main Nextvisit platform.
Audio is never stored. Every account includes a signed BAA.
## What you get
- Live transcription in the browser
- SOAP, DAP, and BIRP system formats, plus team-authored templates
- Note Copilot for plain-English edits
- 100 notes per month
- Copy-out into SimplePractice, TherapyNotes, Jane, Alma, Headway, and other charts
- 14-day trial, no credit card
The product page is [Nextvisit Go](/go). Sign up at [go.nextvisit.ai](https://go.nextvisit.ai/sign-up).
## How it differs from Provider+
Provider+ is part of the main Nextvisit platform. It includes AriaMD, patient timeline, EHR push, ePrescribe, insurance and billing, and supervision. Use it if you prescribe, run a group, or need the chart and claims stack.
Use Nextvisit Go if you need notes and copy-out without that stack.
See [Plans and pricing](/help/plans-and-pricing) for the side-by-side list.
================================================================================
URL: https://nextvisit.ai/help/plans-and-pricing
Title: Plans and pricing
Description: Nextvisit Go, Provider+, and Enterprise. What each includes, monthly vs. annual billing, AI credits, and paid add-ons.
Date: 2026-09-01
Topic: Workspace and billing
================================================================================
Nextvisit offers two separate products. **Nextvisit Go** is for non-prescribing therapists and counselors. The main Nextvisit platform offers **Provider+** and **Enterprise** for prescribers and groups. Every plan includes HIPAA compliance, a signed BAA, and US-hosted data.
## Plans
### Nextvisit Go — $49/month
Nextvisit Go is a separate product for non-prescribing providers, including therapists and counselors. It does not include prescribing or medication-management workflows. Product page: [Nextvisit Go](/go). Sign up at [go.nextvisit.ai](https://go.nextvisit.ai/sign-up).
- SOAP, DAP, and BIRP note formats
- Live transcription
- Note Copilot
- 100 notes per month
- Copy-out into the EHR you already use
- Audio is never stored
- HIPAA and BAA included
- 14-day trial, no card required
Annual billing is $42/month.
### Provider+ — $199/month
For prescribers who run a full patient panel. This is the most popular plan on the full Nextvisit platform, and the one new platform signups start a free trial on.
- AriaMD AI assistant
- Custom templates and cohort tags
- AI-assisted peer review
- AI patient timeline
- Unlimited notes
### Enterprise — custom pricing
For groups, networks, and academic sites. Contact sales for a quote.
- SSO and advanced reporting
- BAA, SOC 2 Type II, and AI governance controls
- Supervision workflows
- Custom templates at scale
- Dedicated Customer Success Manager
## Monthly or annual billing
Every plan bills monthly by default. Switching to annual billing saves 15%. Change your billing cycle from [Billing Portal](/help/billing-portal).
## Free trial and setup
- Start free on Nextvisit Go or Provider+ — no credit card required
- 14-day trial on Nextvisit Go
- Live in two weeks or less, with a BAA signed by default
## AI credits
Credits meter AI work by task complexity, not by note. The AI credits included per billing period depend on plan:
- **Nextvisit Go** — 250 AI credits
- **Provider+** — 1,000 AI credits
- Any other plan priced above $49/month, Enterprise included, defaults to 500 AI credits unless a different amount is quoted
Task complexity tiers:
- **Ultra ($$$$)** — highest cost. Long-context, multi-step reasoning such as longitudinal analysis and batch peer review.
- **Normal ($$$)** — balanced, the default for most automations: notes, event-triggered tasks, and day-to-day prompts.
- **Fast ($$)** — low cost. Simple, single-step tasks such as tagging, classification, and short extractions.
- **Budget ($)** — lowest cost. Lightweight work such as lookups, summaries, and high-volume background tasks.
See remaining credits and where they went on [Usage and credits](/help/usage). For additional credits or volume pricing, contact sales.
## Paid add-ons
Billed per workspace, on top of a plan. These attach to Provider+ and Enterprise. Nextvisit Go is notes and copy-out only.
- **ePrescribe** — Surescripts-certified e-prescribing with EPCS for controlled substances. Per-provider pricing; contact sales for a quote.
- **AriaMD voice agent** — outbound calls for intake, screeners, and insurance capture. Per-call or per-provider pricing; contact sales for a quote.
- **Virtual Fax** — HIPAA-compliant inbound and outbound fax, indexed in the workspace. Pricing varies by volume; contact sales for a quote.
- **Insurance eligibility and claims** — real-time eligibility, claim scrubbing, denial triage, and a full ledger. See [Insurance coding](/help/insurance-coding); contact sales for pricing.
## Changing your plan
View your current plan, browse other plans, and see what each includes from [Billing Portal](/help/billing-portal). Upgrades and downgrades are prorated automatically.
================================================================================
URL: https://nextvisit.ai/help/session-timeouts
Title: Session Timeouts (Auto Log Off)
Description: Why am I being logged out? Understanding automatic session timeouts and how to stay signed in.
Date: 2026-09-01
Topic: Getting started
================================================================================
## Why does Nextvisit AI sign me out automatically?
After **120 minutes** of inactivity, Nextvisit locks the session to protect patient data.
## What happens when the session times out?
**Session Inactivity Notice** asks if you want to keep working. Click **Continue Working** to stay signed in.
If the session locks, the dialog is **Session Locked**. Click **Unlock** to sign in again in a popup window.
## What counts as an inactive session?
* No mouse, keyboard, or touch in the Nextvisit window
* The device screen is locked or asleep
* You have switched to other applications for an extended period
## Tips for a smoother experience
**Install as a Chrome app**
Installing Nextvisit AI as a Chrome app provides better stability and reduces unexpected logouts. Look for the install
icon in your Chrome address bar.
**Check your ad blocker**
Ad blockers can interfere with Nextvisit AI's normal functions even though no ads are displayed. If you experience
frequent disruptions, add Nextvisit AI to your ad blocker's allowlist.
**VPN and proxy considerations**
VPNs or proxy servers may trigger our security firewall and restrict account access. If you experience connection issues
after connecting to a VPN, that may be the cause.
Contact support if unexpected logouts continue.
================================================================================
URL: https://nextvisit.ai/help/what-is-nextvisit-ai
Title: What is Nextvisit AI, and how can it benefit my practice?
Description: An overview of Nextvisit AI for clinical documentation and how it benefits behavioral health practices.
Date: 2026-09-01
Topic: Getting started
================================================================================
Nextvisit AI is an AI scribe built for behavioral health clinicians. It records your patient sessions, then generates
structured clinical notes so you can spend less time on documentation and more time with patients.
Non-prescribing therapists and counselors use [Nextvisit Go](/go), a separate note workspace. Psychiatrists, NPs, and other prescribers use Provider+ or Enterprise on the main Nextvisit platform. See [Plans and pricing](/help/plans-and-pricing).
================================================================================
URL: https://nextvisit.ai/help/user-profile
Title: Managing Your User Profile
Description: Access and customize your account settings, security options, and session management.
Date: 2026-08-31
Topic: Workspace and billing
================================================================================
## Accessing your profile
1. Click your profile icon in the **bottom left corner** of the screen
2. Select **"Profile"** from the dropdown menu
## Profile customization
**Profile image:** Upload a profile picture. It appears throughout the application while you're logged in.
**Display name:** Set how your name appears to other users in shared documents, progress notes, and collaboration
features.
## Account integration
If you signed in with Google or Apple, Profile shows **Signed in with** and a **Google** or **Apple** badge.
## Security settings
**Password:** Update your account password at any time.
**Two-factor authentication (2FA):** Enable 2FA to add an extra layer of security to your account.
## Session management
**Active sessions:** View all devices and browsers where you're currently logged in, including device type, browser, and
last activity.
**Remote logout:** Click **"Log out other browser sessions"** to immediately sign out from all other devices. Useful if
you've used a shared computer or want to confirm no sessions are open elsewhere.
================================================================================
URL: https://nextvisit.ai/help/billing-portal
Title: Accessing Your Billing Portal
Description: How workspace admins open Billing Portal, what it shows, and how Stripe stores card data.
Date: 2026-08-30
Topic: Workspace and billing
================================================================================
Only workspace administrators can access billing settings.
1. Click your profile icon in the bottom left sidebar
2. Select **Billing Portal** from the menu
A valid mobile phone number is required for security when signing up for any plan, including the free tier. VoIP numbers
are not accepted.
## What you'll find
### Notices and messages
Important billing notifications appear at the top of the portal.
### Current plan and available options
- View your current subscription plan
- Browse plans you can switch to
- See what features each plan includes
See [Plans and pricing](/help/plans-and-pricing) for what each plan costs and includes.
### Payment methods
- View saved payment methods and your current default
- Add or update payment methods
### Payment information
Enter your complete billing address for your practice or organization. This appears on your invoices.
### Invoice email settings
Invoices go to your account email by default. You can add additional addresses to receive copies.
## Notes
- All credit card information is stored and processed by Stripe.
- When you upgrade or downgrade, billing is automatically prorated based on your usage.
- Remaining credits and AriaMD spend are on **Usage**. Open **Usage & credits** in the same profile menu, or **Settings → Usage**. See [Usage and credits](/help/usage).
If you can't access the billing portal, confirm you have administrator privileges for your workspace. Open chat at
[help.nextvisit.ai](https://help.nextvisit.ai/docs) for help with billing or subscription changes.
================================================================================
URL: https://nextvisit.ai/help/custom-template-guide
Title: Custom Template Guide
Description: Write a custom note template: placeholders, instruction blocks, examples, and a starter SOAP.
Date: 2026-08-30
Topic: Notes and AI
================================================================================
## How it works
1. You [write a template](https://nextvisit.app/apps/custom-prompts) with instructions and examples
2. The system automatically replaces placeholders (like patient name, age, etc.) before the AI processes the transcript
3. The AI follows your template structure and generates the note
The AI never sees raw placeholder variables. They are swapped with real values first.
Saved templates appear in **Write a note**, **Upload Audio**, **Change Template** → **Custom**, and the recorder **Your Templates**.
If **Mental Status Exam** is on in Note Sections, Nextvisit adds a Mental Status Exam block even when this template has no such section. With **Mental Status Exam** off, the template's mental status section is left as written.
---
## Placeholders
These are automatically replaced by the system before processing. Use them anywhere in your template.
| **Placeholder** | **What It Becomes** | **Notes** |
|-------------------|------------------------------------------------|---------------------------------|
| `[patient_name]` | Patient's name | Always available |
| `[provider_name]` | Provider's name | Always available |
| `[profession]` | Provider profession/taxonomy | Always available |
| `[date]` | Session date (MM/DD/YYYY) | Always available |
| `[dob]` | Patient date of birth | Always available |
| `[scales]` | Recorded scales/assessments from the encounter | Always available |
| `[age]` | Patient's age | May be `-` if no linked patient |
| `[gender]` | Patient's gender/birth sex | May be `-` if no linked patient |
**Handling `-` values:** Some placeholders may contain `-` when patient history isn't available or no patient is linked.
Your template instructions should account for this gracefully (e.g., "If `[age]` is `-`, omit age from the opening
line").
---
## Template structure
Write the template in standard markdown: headers, bold, bullets, tables, and horizontal rules.
```markdown
## Section Header
Content and structure here...
```
---
## Instruction blocks
Instruction blocks tell the AI _how_ to fill in each section. They are written as HTML comments and are never included
in the final note output.
### ``
Tells the AI what to do, what to include, and how to handle missing information.
```markdown
**Chief Complaint:**
```
### ``
Shows the AI a concrete example of what good output looks like for that section.
```markdown
```
Use both. Instructions set the rules. Examples show what following them looks like.
---
## Building a section
Pattern for any section:
```markdown
## Section Name
**Field Name:**
```
### Tips for writing good instructions
* **Be specific about defaults.** If something isn't mentioned in the transcript, tell the AI exactly what to write (
e.g., "If not mentioned, default: Denies headaches or dizziness.")
* **Be specific about what's required vs. optional.** If a section can be omitted when not applicable, say so. If it
must always appear, say that too.
* **Use "never" and "always" deliberately.** The AI takes these literally. "Never use positive statements as the chief
complaint" means exactly that.
* **State the format you want.** Narrative paragraphs? Bullet points? A table? Tell the AI directly.
* **Call out common mistakes.** If you've seen the AI do something you don't want, add a line like: "Do NOT summarize
the transcript directly" or "Avoid generic CBT unless specific techniques are documented."
### Tips for writing good examples
* **Make them realistic.** Use examples that look like actual notes from your practice.
* **Show the format.** If you want bullet points with sub-items, your example should have bullet points with sub-items.
* **Show edge cases when helpful.** An `` block can clarify what you don't want.
---
## Conditional sections
Some sections should only appear under certain conditions. Handle this in your instructions:
```markdown
### Scales and Assessments:
```
```markdown
### Telehealth Documentation:
```
---
## Medication sections
Recommended medication structure:
```markdown
#### Current Medications:
- ****
- **Sig:**
- **Changes:**
- **Adherence:**
- **Side Effects:**
- **Comments:**
```
The `` placeholders inside the bullets show the expected format. The AI replaces them from the transcript.
---
## Tables
Use standard markdown tables for history, diagnostic codes, and review of systems:
```markdown
| ICD-10 Code | Diagnosis |
| :--- | :--- |
| F90.0 | Attention deficit hyperactivity disorder, predominantly inattentive type |
| F33.1 | Major depressive disorder, recurrent, in partial remission |
```
---
## Common patterns
### Required in every note
If a section must always appear regardless of transcript content, state it clearly:
```markdown
```
### Handling missing information
Be explicit about what to do when something isn't discussed:
```markdown
```
### Enforcing consistency across sections
If two sections need to match (e.g., diagnoses in the Assessment must match the Plan), say so:
```markdown
```
---
## Starter template
Minimal starter. Add, remove, or modify sections:
```markdown
## Subjective:
**Chief Complaint:**
**HPI:**
---
### Medications:
#### Current Medications:
- ****
- **Sig:**
- **Changes:**
- **Adherence:**
- **Side Effects:**
- **Comments:**
#### Discontinued Medications:
---
## Objective:
##### Mental Status Examination:
- **Appearance:**
- **Behavior:**
- **Speech:**
- **Mood:**
- **Affect:**
- **Thought Process:**
- **Thought Content:**
- **Cognition:**
- **Insight:**
- **Judgment:**
---
## Assessment:
##### Diagnostic Codes:
---
## Plan:
```
---
## Quick reference
| **Element** | **Syntax** | **Purpose** |
|--------------------|----------------------------------|-----------------------------------------------|
| Instructions | `` | Rules the AI follows for each section |
| Examples | `` | Shows the AI what good output looks like |
| Anti-examples | `` | Shows the AI what NOT to do |
| Placeholders | `[patient_name]`, `[age]`, etc. | Auto-replaced by the system before processing |
| Inline field hints | `` | Shows the AI the expected format for a field |
| Markdown headers | `##`, `###`, `#####` | Structures note sections |
| Horizontal rules | `---` | Visual section separators |
---
## FAQs
**Can I use any markdown formatting?** Yes. Headers, bold, italics, bullet points, numbered lists, tables, and
horizontal rules all work.
**What happens if I don't include an instruction block for a section?** The AI will still attempt to fill the section
based on the transcript, but without specific guidance it may not format or prioritize content the way you prefer.
**Can I have multiple examples for one section?** Yes. You can include multiple `` blocks or put
multiple examples inside a single block, labeled clearly (e.g., "EXAMPLE 1:", "EXAMPLE 2:").
**Do I need both instructions AND examples?** No, but using both produces the best results. Instructions set the rules,
examples show what following those rules looks like in practice.
**Can I make a section conditional?** Yes. State the condition in your instructions (e.g., "Only include this section if
the visit was conducted via telehealth").
**How do I handle placeholders that might be `-`?** Add handling instructions like: "If `[age]` is `-`, omit age from
the opening line" or "If `[scales]` is `-`, omit this section entirely."
**Is there a length limit for templates?** Templates should be as long as they need to be to produce accurate notes, but
excessively long templates with redundant instructions may reduce output quality. Be concise and specific.
================================================================================
URL: https://nextvisit.ai/help/customizing-ai-settings
Title: Customizing Your AI Settings
Description: Configure your AI preferences to match your documentation style and workflow.
Date: 2026-08-30
Topic: Notes and AI
================================================================================
These preferences are per-account, not workspace-wide.
## Accessing your AI settings
Two paths:
1. Click your profile in the bottom left sidebar, then select **AI Preferences**
2. Go to **Settings**, then **AI Preferences**
Settings are configured per user. There is no way to set global preferences for all users in your workspace.
## Healthcare profession
Select your profession from the dropdown. That sets terminology and documentation conventions for the AI.
## Note components
Choose which sections appear in your progress notes. Toggle on or off:
- **Patient history**: Adds chart history to the note
- **Mental status exam**: Adds psychological and cognitive status to generated notes, including intake and custom templates
- **Review of systems**: Adds body-system review
- **Therapeutic intervention**: Adds the treatment plan details
Enabling more sections produces broader documentation. Fewer sections allow for more detail in each included area.
## Preferences
Control which AI-powered automations apply to your workflow. These labels are on the **AI Features** tab:
- **Auto-update Medications**: Automatically update patient medications from session details.
- **Diagnosis Extraction**: Analyze each session to extract diagnoses from the transcript and context.
- **Auto-apply Chart Changes**: Skip the review step: medication and diagnosis changes from your sessions are applied to the patient chart automatically.
- **Review Before Processing**: Hold each session before any processing so you can review and edit the transcript and context first, then send it for processing.
- **Draft Until Reviewed**: Keep each session in a draft state until you review its proposed medication and diagnosis changes.
- **Session Context**: Reference prior sessions for continuity in patient care.
- **Patient Summary**: Generate complexity scores and treatment summaries.
- **Transcription Enhancements**: Clean up and clarify session transcripts before notes are generated.
## Model settings
- **Default model**: Your preferred prompt type (e.g., "SOAP Note (Standard)")
- **Model instructions**: Extra rules such as "Add a telehealth consent notice to the end of each note" or "Include
vitals if mentioned." The system learns from your edits over time.
## Saving your changes
Click **Save** after making any adjustments. Your settings stay active until you change them.
================================================================================
URL: https://nextvisit.ai/help/getting-more-help
Title: Getting more help
Description: What to do when something looks wrong and this help center does not cover it.
Date: 2026-08-30
Topic: Getting started
================================================================================
If something in Nextvisit looks wrong and you cannot find an answer here, open chat at [help.nextvisit.ai](https://help.nextvisit.ai/docs).
Click **Open support chat**. The chat searches the help center first and answers when it can. When the question needs a person, it brings in a member of the team.
On [help.nextvisit.ai/docs](https://help.nextvisit.ai/docs), each article ends with **Was this article helpful?** (Yes or No). **Message us** opens a conversation about that article. In chat, the same ask is **Was this helpful?** with **Ask us about this**.
These questions go to a person:
- Clinical, diagnostic, or medication questions about a specific patient.
- Legal questions, including anything about HIPAA compliance.
- Billing disputes or anything involving a charge you don't recognize.
**Log in** (next to the logo) and **My conversations** open past threads at [help.nextvisit.ai/portal](https://help.nextvisit.ai/portal). Guests receive an email link. Conversation text stays off that page until you log in. **My conversations** is for reading. Reply in **Open support chat**.
A conversation shows where it stands: **Open**, **With the team**, **Waiting for a reply**, **Paused**, or **Resolved**. A person who replies is named on that message. When a request is **Resolved**, the thread asks **How did we do?** (1–5). Send a message in chat to reopen it.
If chat is unavailable, email [hello@nextvisit.ai](mailto:hello@nextvisit.ai).
================================================================================
URL: https://nextvisit.ai/help/new-note-and-upload-audio
Title: Writing a note or uploading a session recording
Description: Start a note directly, or upload an existing audio recording to be transcribed and written up.
Date: 2026-08-30
Topic: Notes and AI
================================================================================
Nextvisit turns a session into a finished note two ways: write it yourself
and let a template shape it, or upload an audio recording and let Nextvisit
transcribe and write it for you. For a live visit, see
[How to record a live session](/help/recording-a-session).
## Writing a note
Click **Write a note** to open the compose window (Classic view: **New Note**), then:
1. Choose the **patient** (or add a new one on the spot).
2. Choose the **provider** the note is for.
3. Choose a **template**. The list includes built-in templates and your custom templates.
4. Write in the **Note** tab. Once a patient is selected, a **Context** tab
appears for background like medications and allergies that should inform
the note but shouldn't be part of the note text itself, and a
**Documents** tab appears if the patient has documents you can link.
Press **⌘/Ctrl + Enter**, or click **Create note**, when you're ready.
## Uploading audio
Click **Upload audio** to open the **Upload Audio** window, then:
1. Choose the **patient** and **provider**, same as above.
2. Choose a **template**. The list includes built-in templates and your custom templates.
3. Drag an audio file onto the upload area, or click it to browse. Supported
formats: MP3, WAV, FLAC, WEBM, OGG, MP4, MOV, and M4A.
4. Optionally add **Context** — medications, allergies, and conditions — the
same way you would for a typed note.
Click **Upload Audio** to submit. Nextvisit transcribes the recording and
generates the note using your chosen template; you don't need to keep the
window open while that happens.
================================================================================
URL: https://nextvisit.ai/help/note-sections-settings
Title: Note Sections Settings
Description: Choose which clinical sections appear in your AI-generated notes. Toggle sections on or off based on your documentation requirements.
Date: 2026-08-30
Topic: Notes and AI
================================================================================
## How it works
1. Navigate to **Preferences -> AI Settings -> Note Sections**
2. Use the toggle switches to enable or disable each section
3. Click **Save Changes**
Enabled sections appear in generated notes. Disabled sections are omitted.
If **Mental Status Exam** is on, Nextvisit adds that section even when the template has no mental status heading.
## Available sections
| **Section** | **Description** | **Default** |
|-------------------------------|--------------------------------------------------|-------------|
| **Patient History** | Relevant medical history background | ON |
| **Mental Status Exam** | Psychological and cognitive status documentation | ON |
| **Review of Systems** | Body systems coverage for thorough documentation | ON |
| **Therapeutic Intervention** | Personalized treatment plan details | ON |
| **Encounter Header & Footer** | Patient info and provider details added to notes | OFF |
Each section is independent. Mental Status Exam and Therapeutic Intervention matter most for behavioral health. Enable
Header & Footer if your EHR does not already add patient and provider information.
================================================================================
URL: https://nextvisit.ai/help/progress-note-templates
Title: Understanding Progress Note Templates
Description: Learn about the different templates available and how to choose the right one for your patient encounters.
Date: 2026-08-30
Topic: Notes and AI
================================================================================
## Available templates
### SOAP Note (Standard)
* **Processing time**: Less than 10 seconds
* **Patient history review**: Essential information only
* **Detail level**: Standard
* **Best for**: Shorter sessions and routine follow-ups
* **Plan**: Available on all plans
### SOAP Note (Advanced)
* **Processing time**: Around 25 seconds
* **Patient history review**: Deep analysis of patient history, timeline of events, and family history
* **Detail level**: Fuller narrative, including history and context
* **Best for**: Behavioral health sessions and encounters up to 2 hours
* **Plan**: Requires a paid plan
### SOAP Note (PRO)
* **Processing time**: Around 1 minute
* **Patient history review**: Full chart history
* **Detail level**: Most detailed, including medications, conditions, and outcomes
* **Best for**: Complex cases that need more detail
* **Plan**: Requires a paid plan; uses an additional 0.5 note credits
### New Patient Intake
* **Processing time**: Similar to Advanced
* **Patient history review**: Deep analysis with intake-specific elements
* **Detail level**: Full intake sections
* **Best for**: First-time patient visits
* **Plan**: Requires a paid plan
Therapists will also see a "DAP Note" option. Additional templates may be available depending on your profession and
account type.
## Changing templates after creation
On an unsigned note, click **Template**. The **Change Template** window opens with **Built-in** and **Custom** tabs.
1. Pick a template
2. Click **Apply Template**
With no custom templates, **Custom** shows **No custom templates yet** and **Create one from Apps → Templates.**
Applying a different template uses an additional note credit.
## Customizing your templates
### Note components
In AI Settings, you can select specific note components to include, such as Therapeutic Intervention.
### AI instructions
Use the AI Instructions area in AI Settings to add custom requirements, for example:
* "Use concise bullet points"
* "Add a statement confirming patient consent"
Use Standard for shorter visits. Advanced or PRO for longer or more complex cases. Advanced and PRO need a paid plan.
PRO uses an additional 0.5 note credits. Remaining credits are on [Usage and credits](/help/usage).
================================================================================
URL: https://nextvisit.ai/help/usage
Title: Usage and credits
Description: See remaining credits, AriaMD spend, and the credit ledger.
Date: 2026-08-30
Topic: Workspace and billing
================================================================================
**Usage** shows remaining credits, AriaMD spend, and a ledger of where credits go. Open it from **Usage & credits** in the profile menu, or from **Settings**.
## Open Usage
1. Click your profile in the bottom left sidebar
2. Select **Usage & credits**
Or open **Settings**, then **Usage**.
On the full AriaMD page, the header shows remaining credits as a number plus **left**. Hover for **Used on AriaMD** and **Credits available**. Click the chip, or choose **View usage and credits** in the command palette.
The sidebar credits meter also opens **Usage**. With the sidebar collapsed, that control is the coins icon.
## What Usage shows
- **Credits available**
- **This billing period** (plan credits used, plan credits remaining, renew date) on a subscription, or **Last 30 days** with no active subscription period
- **Where your credits went**, including **AriaMD Chat**
- **Activity**, with **Added** / **Used** and category filters such as **AriaMD Chat**
The AriaMD chip and **View usage and credits** open **Usage** already filtered to **AriaMD Chat**. The subtitle then reads: AriaMD usage and remaining credits for this billing period.
If another person's subscription covers you, Usage says: "Your usage is covered by [name]'s subscription. You are viewing the shared practice wallet, including activity from other covered members."
Plan, payment methods, and invoices stay on [Billing Portal](/help/billing-portal). See [Plans and pricing](/help/plans-and-pricing) for what each plan costs and how AI credits are metered.
================================================================================
URL: https://nextvisit.ai/help/using-ariamd
Title: Using AriaMD
Description: Open AriaMD from the sidebar, talk with Talk to Aria (Beta), and see remaining credits.
Date: 2026-08-30
Topic: Notes and AI
================================================================================
AriaMD is the assistant in Nextvisit. Open it from the sidebar row **AriaMD**, or press **⌘I** (Mac) / **Ctrl+I** (Windows). The widget menu **Open full Aria** opens the full page at `/chat`.
Aria can make mistakes. Verify clinical information.
## Talk to Aria (Beta)
**Talk to Aria (Beta)** on the composer starts a live voice conversation. Allow the microphone when the browser asks.
## Files in chat
Paperclip → **Upload file** or **Browse files**. Image, PDF, or text. Up to 5 files, 25MB each.
PDF uploads are temporarily limited due to a document processing issue.
Patient-chart PDFs (Documents, 25MB, note credits) are a separate upload. See [Uploading Documents to Patient Charts](/help/uploading-documents).
## Conversation titles
After the first message, the conversation list names the chat. On the full Aria page, open the conversation menu to **Rename** or **Delete**.
## Remaining credits
The AriaMD header shows remaining credits as a number plus **left**. Hover for **Used on AriaMD**. Click the chip, or choose **View usage and credits** in the command palette, to open **Usage** filtered to **AriaMD Chat**.
From other pages: profile menu **Usage & credits**, or **Settings → Usage**. See [Usage and credits](/help/usage).
================================================================================
URL: https://nextvisit.ai/help/activity-log
Title: Activity Log
Description: What the workspace Activity Log records, how to open it, and how to request a CSV export.
Date: 2026-08-29
Topic: Workspace and billing
================================================================================
## How to access the Activity Log
1. Click your **profile icon** in the bottom left sidebar
2. Select **Activity Log** from the popover menu
Credit spend and remaining credits are on **Usage**. Open **Usage & credits** in the same profile menu. See [Usage and credits](/help/usage).
## What is tracked
- **Chart views**: Every time a patient chart is accessed
- **Patient records**: All patient record interactions
- **Sign-in activity**: Successful login attempts
- **Failed sign-ins**: Unsuccessful login attempts
- **API usage**: All API interactions and data exchanges
- **User details**: Who performed each action
- **Timestamps**: When each activity occurred
- **Access details**: What specific information was accessed
## Understanding log entries
Each entry includes the action type, the team member who took the action, a timestamp, and details about what was
accessed or modified. The log records each view of the log as an entry.
## Exporting activity data
To request a CSV export for compliance purposes or deeper analysis:
1. Contact Nextvisit support
2. Provide your **workspace name**
3. Specify the **date range** you need
4. Allow 2-3 business days for processing
Use the log for audits, unusual access patterns, failed logins, and incident review.
================================================================================
URL: https://nextvisit.ai/help/edit-a-note-with-aria
Title: Edit a note with Aria
Description: Ask Aria to change an unsigned encounter note, preview the result, then apply or discard.
Date: 2026-08-29
Topic: Notes and AI
================================================================================
On an unsigned encounter note, the composer under the note says **Tell Aria what to change…**. Focus the row to expand it.
Type an instruction, or pick a starter: Tighten the wording, Expand the plan, Add a mental status exam, Fix grammar and flow.
Aria keeps the note on the page. A preview appears when there are edits. Switch **Note** and **Changes**. **Apply to note** writes the draft. **Discard** drops it.
Signed notes hide this composer. See [How to View and Restore Progress Note Version History](/help/note-version-history).
================================================================================
URL: https://nextvisit.ai/help/insurance-coding
Title: How to Set Up and Use Insurance Coding
Description: Add billing codes to encounters so they shape the note and appear on exported PDFs.
Date: 2026-08-29
Topic: Workspace and billing
================================================================================
Billing codes such as CPT 90833 track services on the encounter. The AI uses the code to shape the note. Billing codes on the encounter print on the PDF as **Service Codes**. Practice name and address come from workspace settings.
## How to access billing code settings
1. Click your profile icon in the bottom left of the sidebar
2. Select **Settings** (gear icon)
3. Click **Insurance Billing**
## Setting up your billing codes
From the Insurance Billing settings page, you can create and manage your billing code library. Each code requires:
- **Code** (required): The billing code number (e.g., 90833)
- **Modifier** (optional): Any applicable modifiers
- **Price**: The fee associated with this code
- **Duration**: Expected session length (e.g., "15 Minutes")
- **Description** (required): A description of the service
Once saved, codes are available across all your patient encounters.
## Adding billing codes to encounters
**During an active session:** Select the appropriate billing code while recording to guide documentation in real time.
**After the encounter:** Add the billing code when reviewing an unsigned encounter. A signed encounter shows **Note Signed**. Click **Unsign** to change codes. Submitted insurance claims that are not drafts hide **Unsign** and the coding **Add** / **Edit** control.
Complete practice information in workspace settings so the letterhead prints with the codes.
## AI coding suggestions
When Nextvisit's AI identifies a billing code that fits the documented session and meets the code's criteria, it appears
under **suggestions** with a confidence percentage. Click **Show facts** on a suggestion to see the specific parts of
the note that support it.
- **Apply** — adds the suggested code to the encounter's billing codes.
- **Dismiss** — removes the suggestion. It stays dismissed for that encounter unless the AI regenerates new suggestions.
The system does not apply suggestions automatically. Review each suggestion. Apply or dismiss it.
================================================================================
URL: https://nextvisit.ai/help/note-version-history
Title: How to View and Restore Progress Note Version History
Description: Open version history on an unsigned note, preview a prior save, and restore it.
Date: 2026-08-29
Topic: Notes and AI
================================================================================
Nextvisit saves version history automatically.
On an unsigned note, click **Change History** (clock-arrow icon). The **Version History** panel opens. Preview a prior save, then click **Restore Version**.
A signed note shows **Note Signed**. Change History and Edit are hidden. Click **Unsign**, then restore. Submitted insurance claims that are not drafts hide **Unsign**. The product shows: `Cannot unsign when submitted claims are present`.
## Accessing version history
1. Open the progress note you want to review
2. Click **Change History**
3. The **Version History** panel opens
## What the panel shows
* Each version is listed with the date and exact time it was saved
* The most recent version is marked "Latest"
* Click any version to preview its content
* **Restore Version** applies to the version you selected
## Restoring a previous version
1. Browse the versions listed in the panel
2. Click a version to preview it
3. Click **"Restore Version"** at the bottom of the panel
4. Confirm, or click **"Cancel"** to go back
Restoring replaces the current note text. The restored version becomes the new latest. All prior versions stay in the list.
================================================================================
URL: https://nextvisit.ai/help/patient-medications
Title: Managing Patient Medications
Description: Where to find the medication list, what you can edit, and the status values (Current, Titration, Tapering).
Date: 2026-08-29
Topic: Patient charts
================================================================================
After a session, Nextvisit can propose medication and diagnosis changes. The **Review proposed chart changes** dialog lets you **Accept**, **Reject**, or **Modify** then **Apply changes**.
**Auto-apply Chart Changes** in AI Features applies those proposals without the review step.
If you Accept a change whose chart row was removed, the product shows: `This chart entry is no longer on the patient chart.` That proposal stays in the list.
## What you can do
After completing at least one session with a patient, you can:
* Add new medications
* Edit existing medications (dosage, frequency, status)
* Delete medications that are no longer relevant
* Print the complete medication list for patient records or referrals
* Switch an agent with **Switch** / **Switch agent**
## Where to find patient medications
1. **During an encounter or progress note:** Look for the "Medications" section in the right sidebar when medications
were discussed in the session.
2. **In the patient chart:** Navigate to the patient's chart and click the "Medications" tab for a full view.
## Switch agent
On the Medications tab, **Switch** opens **Switch agent**. A row control labeled **Switch agent** opens the same
dialog with that medication selected.
1. Choose **Switch from** and **Switch to**
2. Click **Propose schedule**
3. **Accept**, **Modify** (**Save modified plan**), or **Reject** (**Reject plan**, with a reason)
After **Accept** or **Modify**, the product shows: `Switch accepted; outcomes are due on day 30 and day 90.` (or
`Modified plan saved; outcomes are due on day 30 and day 90.`)
When those dates arrive, **Switch outcomes due** offers **Record**. The dialog title is **Day 30 outcome** or
**Day 90 outcome**.
## Editing medication details
Click any medication to open a detailed view that includes:
* **Medication name** and generic equivalent
* **Dosage** (e.g., 10mg)
* **Formation** (tablet, capsule, liquid, etc.)
* **Frequency** (once daily, twice daily, as needed)
* **Start date** and **stop date** (if applicable)
* **Status** options including Current, Discontinued, Titration, Tapering, or Unknown
* **Prescriber information**
* **Comments section** for additional notes like "Take for sleep before bedtime"
* **Drug interactions** and warnings when applicable
## Medication status options
* **Current**: Patient is actively taking the medication
* **Discontinued**: Medication has been stopped
* **Titration**: Dosage is being gradually increased
* **Tapering**: Dosage is being gradually decreased
* **Unknown**: Status needs clarification
## AI confidence indicators
Nextvisit AI shows confidence levels for automatically captured medications. Look for confidence percentages and
controlled substance indicators where applicable.
## Notes
* At least one completed session is required before you can manage a patient's medications
* Review and confirm medication details before finalizing documentation
* Use "Mark Incorrect & Remove" if the AI captured incorrect information
================================================================================
URL: https://nextvisit.ai/help/practice-fusion-integration
Title: Practice Fusion Integration
Description: Practice Fusion beta: SMART launch, what it pulls today, setup fields, and how to copy notes back.
Date: 2026-08-29
Topic: EHR and integrations
================================================================================
## Getting started
Nextvisit AI integrates directly with Practice Fusion. The integration is currently in Beta and available to all paid
plan users at no additional cost.
## What's supported
**Current:**
* **Patient information access** - Pull patient demographics and encounter details automatically
* **SMART launch** - Open Nextvisit AI directly from a patient encounter in Practice Fusion
**Coming soon:**
* **Direct note writing** - Completed notes will save automatically to Practice Fusion (in development)
## How to use the integration
Once set up:
1. Open a patient encounter in Practice Fusion
2. Launch Nextvisit AI from the patient view
3. Complete your documentation as usual
4. Copy your finished notes back to Practice Fusion using the copy button on the note, or right-click a progress note in
your list and select "Copy..." then "Note"
## Setting up your integration
Setup requires a brief process on our end. To start:
1. Open chat at [help.nextvisit.ai](https://help.nextvisit.ai/docs)
2. Provide the following from your Practice Fusion account:
* Practice ID
* Practice name
* Practice address
* Practice phone number
The integration is officially supported by Practice Fusion and uses FHIR. For setup questions, open chat at
[help.nextvisit.ai](https://help.nextvisit.ai/docs).
================================================================================
URL: https://nextvisit.ai/help/workspace-settings
Title: Workspace Settings
Description: Configure your practice information, workspace details, and team members.
Date: 2026-08-29
Topic: Workspace and billing
================================================================================
## Practice information
This information appears on your PDF exports.
**Practice Legal Name:** The official name of your practice or facility as it appears on your professional documents.
**Address:** Your practice's full or partial address.
The first page of an encounter PDF lists **Provider** (the documenting clinician), **Date**, **Patient Name**, and **Date of Birth** when the chart has a date of birth.
Every page prints the patient name at the top. With a date of birth on the chart, that line is the patient name, a middle dot, and `DOB MM/DD/YYYY`.
A signed note prints `Electronically signed by [clinician] on [date and time]`.
## Workspace configuration
**Workspace Name:** A short, descriptive name such as your practice name, a provider's name, a location, or a
department.
**Members:** Open **Settings → Workspace → Members**. The heading is **Clinic Members**.
**Add Clinic User** adds a person with name, email, and a role (**Provider** or **Staff**). Confirm with **Add member**.
On **Clinic Members**:
- **Archive** (another active member) opens **Archive Clinic Member**: "Archive this person to revoke clinic access while keeping their notes and records. You can restore them later." Confirm **Archive**. The row stays with an **Archived** badge. Notes stay attributed to them.
- **Restore** (an **Archived** row) opens **Restore Clinic Member**: "Restore this person to the clinic as staff? They will regain access immediately." Confirm **Restore**. They return as **Staff**.
- **Remove** (another active member) opens **Remove Clinic Member**: "Are you sure you would like to remove this person from the clinic?" Confirm **Remove**.
- **Leave** (your own active row) opens **Leave Clinic**: "Are you sure you would like to leave this clinic?" Confirm **Leave**.
An archived person who signs in sees: `Your account has been archived. Please contact your administrator for access.`
**Billing note:** Users with the "Provider" role are billable, and billing is prorated when added. Only assign this role
to users who will actively document patient encounters. Archiving a **Provider** drops that billable seat. Archived
members are left out of the team user count.
================================================================================
URL: https://nextvisit.ai/help/your-dashboard
Title: Your dashboard
Description: Last time, overdue notes, aging, and starting a visit from Dashboard.
Date: 2026-08-29
Topic: Getting started
================================================================================
Open **Dashboard**. The heading is a greeting with your first name.
When unsigned notes are waiting, the line under the greeting shows **overdue notes** (48 hours or more), **aging** (24 hours or more), or **notes to review**. **overdue notes** opens Sessions with Status **Overdue**. **aging** and **notes to review** open Sessions.
**Pick up where you left off** lists recent sessions. Unsigned notes can show **Overdue** or **Aging**. **All sessions** opens the full list. Sessions can filter Status **Overdue** and badge rows **Overdue** or **Aging**.
**Find a patient (optional)** scopes the visit. **Last time** shows the last note, homework, and chart facts. **First session** shows when there is no prior note. **Last visit is still writing** shows while that note is still processing. **Open last note** opens the session. When the last visit is old, Last time shows **It's been a while.** and hides homework.
**Start recording**, **Write a note**, and **Upload audio** start a visit for the selected patient, or with no patient selected.
**Classic view** switches to the widgets layout.
================================================================================
URL: https://nextvisit.ai/help/ai-settings-overview
Title: AI Settings Overview
Description: Six AI Settings tabs: Profession, Note Sections, Verbosity, Voice Style, AI Features, and Note Template.
Date: 2026-01-28
Topic: Notes and AI
================================================================================
## Accessing AI settings
**Location:** Preferences -> [AI Settings](https://nextvisit.app/settings/ai)
## Settings categories
AI Settings are organized into six tabs:
| **Tab** | **What It Controls** |
|-------------------|------------------------------------------------------------------|
| **Profession** | Adjusts terminology and formatting for your healthcare specialty |
| **Note Sections** | Choose which clinical sections appear in your notes |
| **Verbosity** | Control how detailed your notes are |
| **Voice Style** | Set the writing tone and structure |
| **AI Features** | Enable automated features like medication updates |
| **Note Template** | Select note format and add custom instructions |
## Default configuration
Nextvisit AI comes pre-configured for behavioral health documentation:
* **Profession:** Psychiatrist
* **Note Sections:** All enabled except Header/Footer
* **Verbosity:** Standard
* **Voice Style:** Naturalistic
* **AI Features:** All enabled
* **Note Template:** SOAP Note (Standard)
## Saving your changes
* Each tab has its own **Save Changes** button at the bottom
* A success confirmation appears at the top of the page when saved
* Settings persist across browser sessions and devices
* Changes apply to all future sessions (existing notes are not modified)
If settings are not applying: confirm you clicked **Save Changes**, refresh to verify, and check you are in the correct
workspace. To restore defaults, contact support or reselect the values above.
================================================================================
URL: https://nextvisit.ai/help/custom-instructions
Title: Custom Instructions Settings
Description: Add standing instructions on Preferences -> AI Settings -> Note Template. Examples for consent, ICD-10, and billing.
Date: 2026-01-28
Topic: Notes and AI
================================================================================
## How it works
1. Navigate to **Preferences -> AI Settings -> Note Template**
2. Enter your instructions in the rich-text editor
3. Click **Save Changes**
Your instructions are applied to all future note generation automatically.
## What you can do with custom instructions
Be specific. Examples:
**Adding standard elements:**
* "Add telehealth consent notice to every note"
* "Include vitals if mentioned during the session"
* "Always document time spent on medical decision-making"
**Workflow-specific requirements:**
* "Always note if patient arrived late"
* "Include treatment plan review at end of each note"
* "Add safety assessment section for high-risk patients"
* "Note when interpreter services were used"
**Formatting preferences:**
* "Use ICD-10 codes when documenting diagnoses"
* "Spell out medication names fully, no abbreviations"
* "Include both generic and brand names for medications"
**Compliance and billing:**
* "Document medical necessity for prescribed medications"
* "Include time-based billing elements for E/M coding"
* "Note coordination of care with other providers"
## Editor features
The rich-text editor supports:
* Bold and italic text
* Headers
* Bulleted and numbered lists
* Links
* Tables
* Undo/redo
## How the AI learns from you
Custom instructions are automatically optimized based on your feedback over time. When you consistently edit generated
notes in the same way, the system learns those patterns and incorporates them into future note generation.
### Tips
* Be specific and direct (e.g., "Include X" rather than "You might want to add X")
* Instructions work alongside your other settings (verbosity, voice style, note sections)
* Test new instructions on a few sessions to make sure they produce the results you want
* Update instructions anytime as your documentation needs change
================================================================================
URL: https://nextvisit.ai/help/profession-settings
Title: Profession Settings
Description: Set Profession under AI Settings. Behavioral-health shortcuts plus the full searchable list.
Date: 2026-01-28
Topic: Notes and AI
================================================================================
## How it works
1. Navigate to **Preferences -> [AI Settings](https://nextvisit.app/settings/ai) -> Profession**
2. Select your healthcare profession from the list
3. Click **Save Changes**
The AI adjusts clinical language and documentation conventions to match the specialty.
## Quick select: behavioral health
The most common behavioral health professions appear at the top for easy access:
* **Psychiatrist** - Mental health diagnosis and medication management
* **Nurse Practitioner** - Advanced practice nursing with prescriptive authority
* **Psychologist** - Psychological assessment and therapy
* **Therapist** - Counseling and psychotherapy services
* **Social Worker** - Clinical social work and case management
## All available professions
Use the search bar to quickly find your profession from the complete list:
* **A-D:** Acupuncturist, Anesthesiologist, Audiologist, Cardiovascular Technologist, Chiropractor, Dental Hygienist,
Dentist, Dietitian/Nutritionist
* **E-M:** Emergency Medical Technician (EMT), Genetic Counselor, Home Health Aide, Massage Therapist, Medical
Assistant, Medical Laboratory Scientist, Medical Transcriptionist
* **N-P:** Occupational Therapist, Optometrist, Orthodontist, Orthotist/Prosthetist, Osteopathic Physician, Paramedic,
Pediatrician, Pharmacist, Pharmacy Technician, Phlebotomist, Physical Therapist, Physician, Physician (Medication
Assisted Treatment), Physician Assistant, Podiatrist
* **R-Z:** Radiologic Technologist, Registered Nurse (RN), Respiratory Therapist, Speech-Language Pathologist, Surgeon,
Surgical Technologist, Veterinarian, Veterinary Technician/Technologist
## Tips
* Only one profession can be selected at a time.
* Your selection persists across all future sessions.
* If your exact title isn't listed, choose the closest match to your scope of practice.
================================================================================
URL: https://nextvisit.ai/help/verbosity-settings
Title: Verbosity Settings
Description: Control how detailed generated notes are, from short bullets to a full narrative.
Date: 2026-01-28
Topic: Notes and AI
================================================================================
## How it works
1. Navigate to **Preferences -> AI Settings -> Verbosity**
2. Select your preferred verbosity level
3. Preview the sample text to see how each level looks
4. Click **Save Changes**
## Verbosity levels
### Brief
**Best for:** Quick encounters, follow-up visits, routine appointments
Minimal detail with key points only. Gets straight to the essentials.
**Example:**
_Routine medication refills. Patient reports Vyvanse is less effective at current dose and requests increase. Wants to
discontinue duloxetine (Cymbalta). HPI: Patient takes Vyvanse 40mg in the early morning (7-8 AM) and Adderall XR later
in the day for ADHD..._
---
### Concise
**Best for:** Standard documentation, balanced approach
Essential information with moderate detail. A middle ground for most routine clinical encounters.
---
### Standard (Default)
**Best for:** Detailed clinical notes
Balanced level of detail that covers all relevant clinical information without excessive length.
**Example:**
_Chief Complaint: The patient presents for routine medication refills, specifically for ADHD management. They also wish
to discuss the effectiveness of their current stimulant regimen and request discontinuation of duloxetine (Cymbalta),
which was prescribed by another provider. History of Present Illness: The patient's ADHD is currently managed with a
combination of Vyvanse and Adderall XR..._
---
### Comprehensive
**Best for:** Complex cases, new patient intakes, detailed psychiatric evaluations
Maximum detail and documentation. Use when thorough documentation is required for medical-legal purposes or complex
treatment planning.
The preview updates when you switch levels. You can change verbosity at any time. Use Comprehensive for new patients or
complex cases.
================================================================================
URL: https://nextvisit.ai/help/voice-style-settings
Title: Voice Style Settings
Description: Set Voice Style under AI Settings: Formal, Structured, Naturalistic, Narrative, or Shorthand.
Date: 2026-01-28
Topic: Notes and AI
================================================================================
## How it works
1. Navigate to **Preferences -> AI Settings -> Voice Style**
2. Select your preferred writing style
3. Preview the sample text to see how each style reads
4. Click **Save Changes**
## Voice style options
### Formal
**Best for:** Traditional medical documentation, academic settings
Polished, highly professional clinical language with traditional medical documentation conventions.
#### Formal Example
_The patient presents for routine medication management, specifically regarding their current ADHD treatment regimen and
a request to discontinue duloxetine. History of Present Illness: The patient's attention-deficit/hyperactivity disorder
is currently managed with a combination of lisdexamfetamine (Vyvanse) 40 milligrams and extended-release amphetamine
salts (Adderall XR)..._
---
### Structured
**Best for:** Balance between formality and readability
Organized with clear sections, professional but more approachable than Formal style.
---
### Naturalistic (Default)
**Best for:** Most clinical documentation
Flows like natural conversation with a real-time documentation feel. Professional but not stiff.
#### Naturalistic Example
_The patient presents for routine medication management, specifically regarding their current ADHD treatment regimen and
a request to discontinue duloxetine. The patient's attention-deficit/hyperactivity disorder is currently managed with a
combination of lisdexamfetamine (Vyvanse) 40 milligrams and extended-release amphetamine salts (Adderall XR)..._
---
### Narrative
**Best for:** Providers who prefer story-like documentation
Fluid prose that tells a story and connects information cohesively.
---
### Shorthand
**Best for:** Quick reference, abbreviated documentation
Clinical abbreviations and bulleted lists for maximum efficiency.
#### Shorthand Example
_CC:_
* _Routine med refills_
* _ADHD med effectiveness concerns_
* _Requests d/c duloxetine (Cymbalta)_
* _Requests Vyvanse dose adjustment_
_HPI:_
* _Currently on Vyvanse 40mg + Adderall XR for ADHD_
* _Vyvanse less effective at current dose, requests bump to 50mg_
* _Takes Vyvanse 7-8 AM, Adderall later in day_
The preview uses the same clinical scenario across styles. Shorthand pairs well with Brief verbosity.
================================================================================
URL: https://nextvisit.ai/help/service-locations
Title: Service Locations
Description: Track where patient encounters take place with customizable location settings.
Date: 2025-07-11
Topic: Patient charts
================================================================================
The selected location appears in the encounter sidebar and on PDF exports.
## Activating service locations
The Service Location feature is a free app. To enable it:
1. Click the **shortcuts icon** (three stacked squares) in the top right corner
2. Select **Apps** from the menu
3. Find **Service Location** in the list
4. Click **Activate**
After activation, you'll be redirected to the service locations management page to start adding locations.
## Adding location information
When creating a service location, you can include these details:
* **Name** - The primary location name
* **Alias** - An alternative or shortened name
* **Address** - Street address
* **City** - City name
* **State** - State or province
* **Postal Code** - ZIP or postal code
* **Phone** - Primary contact number
* **Fax** - Fax number (if applicable)
## Managing your service locations
To access service locations after initial setup:
1. Click your **profile picture** in the bottom left corner
2. Select **Settings**
3. Choose **Service Locations** from the settings menu
From this page, add new locations, edit existing ones, or remove ones you no longer need.
## Notes
* Once a location is selected for an encounter, it appears in the right sidebar and on PDF exports.
* Service locations are available to all users in the workspace.
================================================================================
URL: https://nextvisit.ai/help/ai-peer-review
Title: AI Peer Review
Description: Open Peer Review on a completed note for strengths, gaps, treatment options, and impacting factors.
Date: 2025-07-09
Topic: Notes and AI
================================================================================
## How to access AI Peer Review
1. Navigate to a completed progress note or patient encounter
2. Click the **Peer Review** tab at the top of the document
3. The AI will analyze your documentation and generate the review
## What the review covers
**Strengths**
- What you did well in the encounter
- Effective clinical decision-making
- Thorough documentation practices
**Areas to improve**
- Gaps in documentation or assessment
- Areas where additional detail would help
- Potential gaps in clinical reasoning
**Treatment recommendations**
- Alternative treatment approaches to consider
- Evidence-based interventions
- Additional therapeutic options
**Potential impacting factors**
- Patient factors that may influence outcomes
- Social determinants of health
- Comorbidities or risk factors that warrant attention
**Additional recommendations**
- Supplementary clinical insights
- Follow-up considerations
- Preventive care recommendations
## What affects the review
The feedback varies based on the type of session (initial consultation, follow-up, specialty visit), patient complexity,
and how complete your documentation is.
Use it on complex cases, before a significant decision, or when you want a second look at treatment options.
================================================================================
URL: https://nextvisit.ai/help/ai-timeline
Title: AI Timeline: Track Patient Progress at a Glance
Description: View key life events and treatment milestones automatically organized from your patient encounters.
Date: 2025-07-09
Topic: Patient charts
================================================================================
## Where to find the AI Timeline
**In the patient chart:**
- Navigate to your patient's chart
- Click the **Timeline** or **AI Timeline** tab
**During encounters:**
- Look for the timeline widget in the right sidebar while documenting your progress note
## How the timeline works
The AI identifies and displays 3-7 key events per session depending on encounter complexity. Each event includes:
- An emoji that represents the event type
- A brief, readable summary of what happened
- Session grouping for easy navigation
## Getting more details
When viewing the timeline from the chart or patient view, click any event to see:
- **Relevant history**: Background context related to the event
- **Significant details**: Important clinical information and observations
- **Changes**: Updates to prescriptions, condition status, or other treatment modifications
- **Patient quotes**: Direct quotes from the session when applicable and clinically relevant
================================================================================
URL: https://nextvisit.ai/help/athena-integration
Title: Athena Integration
Description: What Athena syncs (meds, conditions, labs), $49/month per clinician, and the authorization steps.
Date: 2025-07-09
Topic: EHR and integrations
================================================================================
## What the integration pulls
- Current medications and dosages
- Patient conditions and diagnoses
- Recent lab work and results
- Other relevant clinical data from the patient's chart
## Pricing
The Athena integration is $49/month per clinician. Group practices with more than 5 users can contact us for custom
pricing.
## Getting started
1. Contact our team to initiate the integration process
2. Complete the authorization form
at: [athenahealth.com/authorization-consent](https://marketplace.athenahealth.com/authorization-consent?product=nextvisit-inc)
3. Wait for confirmation that your integration is active
Once active, Nextvisit pulls relevant patient data from Athena for each encounter. Contact support if you have
questions about setup or authorization.
================================================================================
URL: https://nextvisit.ai/help/audio-storage
Title: How Long Is Recorded Audio Stored?
Description: Understanding audio storage duration and security for your clinical documentation.
Date: 2025-07-09
Topic: Security and data
================================================================================
Storage duration depends on which recording method you use.
## Session recorder and mobile app
Audio is stored for **30 days** from the date of recording.
## Live transcription mode
Audio is **not saved**. Transcription happens in real time with no storage.
## Downloading audio files
To keep a recording beyond 30 days:
1. Navigate to the session
2. Click the **Audio** tab
3. Select the download option
Once you download a file, you are responsible for its security and compliance with local laws, regulations, and your
organization's data policies. Nextvisit cannot track or manage downloaded files.
## Security
All stored recordings are protected with AES-256-CBC encryption.
## Best practices
- Download important recordings before the 30-day window closes, only if you need them long-term
- Store downloaded files according to your organization's data retention policies
================================================================================
URL: https://nextvisit.ai/help/context-area
Title: How to Use the Context Area to Enhance Your Progress Notes
Description: Add background information and private notes to improve your clinical documentation quality.
Date: 2025-07-09
Topic: Notes and AI
================================================================================
## How to use the Context area
1. Open the Context field in the Nextvisit AI interface before or during your patient encounter
2. Add relevant information such as:
- Current medications and dosages
- Known allergies and reactions
- Relevant medical history
- Family history details
- Previous treatment responses
- Clinical observations
- Private notes about the patient's condition
3. Continue adding information throughout the encounter as new details emerge
When using the recorder, you can type in the Context area without speaking information aloud. This is useful for
sensitive observations or anything you want documented privately.
## Pre-filled context
The Context field may already contain information pulled automatically from system integrations, including data from
previous encounters, medication lists, allergy records, and EHR integration data. Review this for accuracy and add
anything it's missing.
## Example entries
- "Patient has been non-compliant with metformin 500mg BID for past 2 weeks"
- "Family history of CAD - father had MI at age 52"
- "Patient appears anxious about upcoming procedure"
- "Previous adverse reaction to penicillin - hives and swelling"
================================================================================
URL: https://nextvisit.ai/help/epic-integration
Title: Epic EHR Integration
Description: Epic integration is in development. Until then, copy a Nextvisit note and paste it into Epic.
Date: 2025-07-09
Topic: EHR and integrations
================================================================================
## Current status
Nextvisit AI is actively developing an Epic integration. No specific completion date is available yet.
## Using Nextvisit AI with Epic today
Until the full integration is available:
1. Complete your patient encounter documentation in Nextvisit AI
2. Click the **Copy** button on your progress note
3. Navigate to the appropriate field in Epic
4. Paste the content directly
The formatting pastes cleanly into Epic. Contact support if you want to join the Epic integration waitlist.
================================================================================
URL: https://nextvisit.ai/help/note-feedback
Title: How to Leave Note Feedback
Description: Help improve your clinical documentation by rating and reviewing your progress notes.
Date: 2025-07-09
Topic: Notes and AI
================================================================================
## Rating your note
After a progress note is generated, you'll see a feedback section at the top of the note with thumbs up and thumbs down
icons.
1. Click **thumbs up** or **thumbs down** to indicate your overall satisfaction
2. Select a rating from **1 to 10** (1 = Awful, 10 = Amazing)
3. Click **Save Feedback**
## Adding detailed feedback (optional)
Select from predefined categories to explain your rating.
**For positive ratings:**
- Accurate medical information
- Clear patient communication
- Evidence-based guidance
- Appropriate empathy
- Thorough clinical explanation
- Proper medical terminology
- Followed clinical guidelines
- Other
**For negative ratings:**
- Inaccurate medical information
- Unclear patient communication
- Lack of evidence-based support
- Inappropriate tone
- Incomplete clinical explanation
- Missed important details
- Exceeded scope of practice
- Other
## Adding comments
After selecting a category, you can leave additional comments explaining what influenced your rating.
Ratings and comments teach the system your style. The more you leave, the closer later notes get.
================================================================================
URL: https://nextvisit.ai/help/practice-analytics
Title: Understanding Your Practice Analytics
Description: Analytics widgets: top medications, conditions, Rx class, treatment status, and top users.
Date: 2025-07-09
Topic: Patient charts
================================================================================
This dashboard is being updated. Current widgets:
## Top medications
Displays your 5 most frequently prescribed medications from the past 30 days. Each entry shows:
* Medication name
* Generic name
* Drug class
* Total prescriptions
Click "Show More" to view all medications detected or prescribed during this period. Select any medication to access
live clinical insights powered by AriaMD.
## Top conditions
Shows your 5 most common patient conditions, including:
* Condition name
* ICD-10 code
* Description
* Total occurrences
Use "Show More" to see your full condition list from the past 30 days. Click any condition for detailed insights from
AriaMD.
## Top Rx class
A pie chart showing the distribution of medication types in your practice (SSRIs, stimulants, and other drug classes).
Useful for reviewing prescribing patterns across therapeutic categories.
## Treatment status overview
AI-estimated breakdown of how your patients are progressing. Categories include Stable, Inconsistent, Declining, and
others. Gives a quick snapshot of overall patient outcomes.
## Top users
Shows which team members are using Nextvisit AI most frequently, useful for tracking adoption across your practice.
================================================================================
URL: https://nextvisit.ai/help/referral-program
Title: Share Nextvisit AI and Earn Rewards
Description: Refer colleagues to Nextvisit AI and earn up to $100 for each successful referral through our referral program.
Date: 2025-07-09
Topic: Workspace and billing
================================================================================
Earn up to $100 when a colleague becomes an active user through your link. Open **"Refer a Friend"** on the dashboard.
## How it works
1. **Share your link:** Every user gets a unique referral link. Copy it directly or customize it before sharing.
2. **They sign up:** When a colleague clicks your link and creates an account, the system tracks the referral
automatically.
3. **Earn rewards:** Once your referred colleague becomes an active user, you receive the reward payment via Stripe.
## Managing your referral link
**Customizing your link**
Tap "Customize Link" to update your display name (shown when someone visits your link) and your referral code (the
identifier in your URL).
**Copying and sharing**
Use "Copy Link" to copy your referral URL to the clipboard, then share it by email, text, or any other method.
## Tracking your referrals
The referral dashboard shows:
* **People referred:** Total colleagues signed up through your link
* **Rewards earned:** Total amount earned from successful referrals
Under Recent Activity, you can see names, sign-up dates, and current status (for example, "Pending Subscription") for
each referral.
## Payout information
Rewards are processed through Stripe.
**Program restrictions**
* Payouts are limited to eligible participants within the United States
* Recipients must provide a valid individual Tax Identification Number (TIN) or business Employer Identification
Number (EIN)
* All referrals must comply with applicable healthcare industry regulations
* Nextvisit AI reserves the right to verify referral eligibility and may request additional documentation before
processing payments
* Program terms and reward amounts are subject to change with notice
================================================================================
URL: https://nextvisit.ai/help/tags
Title: How to Create and Use Tags for Patients and Encounters
Description: Organize your patients and encounters with custom tags for better workflow management.
Date: 2025-07-09
Topic: Patient charts
================================================================================
## Creating a tag
1. Click your profile icon in the bottom left corner, then select "Settings"
2. Click "Tags" in the Settings menu
3. Click "Create new Tag"
4. Configure the tag:
* **Tag Name**: The display name shown on the tag
* **Description**: Your internal reference note (not visible on the tag itself)
* **Color** and **Shade**: Choose from the available palette
5. Click "Create" to save
## Assigning tags to patients and encounters
1. Open the patient chart or encounter you want to tag
2. Click the tag icon
3. Select the tags to assign
## Filtering by tags
1. In your patient list or encounters view, click "Filters"
2. Select "Tags" from the filter options
3. Choose the tag(s) to filter by
## Notes
* Tags are shared across your workspace and available to all workspace members.
* Tags only exist in the workspace where they were created.
================================================================================
URL: https://nextvisit.ai/help/vulnerability-disclosure
Title: Vulnerability Disclosure Program Guidelines
Description: What we're looking for and compensation details for security researchers.
Date: 2025-07-09
Topic: Security and data
================================================================================
Nextvisit AI welcomes responsible disclosure of security vulnerabilities. To focus the program on meaningful security
issues, we've established clear guidelines on what qualifies.
## Compensation policy
We do not currently provide monetary compensation for vulnerability reports. If this policy changes, we will update this
page.
## What we don't accept
The following findings do not qualify for our vulnerability disclosure program:
## Infrastructure and configuration issues
* Missing security headers (CSP, X-Frame-Options, etc.) without demonstrable impact
* SSL/TLS configuration issues, expired certificates, or support for older protocols
* Missing HSTS headers
* Software version disclosure or banner grabbing
## Low-impact or theoretical issues
* Scanner output without manual verification and proof of exploitation
* Theoretical vulnerabilities lacking working proof-of-concept
* Clickjacking on non-sensitive pages
* Cookie flags (Secure/HttpOnly) without security impact
* Rate limiting observations without actual impact
* CORS misconfigurations on public resources
* Open redirects without demonstrable harm
* Directory listings of non-sensitive files
## User interaction and social engineering
* Self-XSS or vulnerabilities requiring victim interaction
* Social engineering attempts
* Issues resulting from user error (exposed API keys, weak passwords)
* AutoComplete/password manager behavior
## Authentication and access issues
* Email spoofing (SPF/DKIM/DMARC configuration)
* Brute force, password spraying, or credential stuffing
* Testing against accounts you didn't create
* Username/email enumeration via timing attacks
* Logout CSRF
## Out-of-scope issues
* Subdomain takeover on out-of-scope domains
* Issues requiring pre-compromised devices/networks
## What we want to hear about
If your finding demonstrates real, exploitable risk to Nextvisit AI (nextvisit.app) systems or user data that isn't in
the exclusion list above, we want to hear about it.
We're looking for vulnerabilities that could:
* Compromise patient data or protected health information
* Allow unauthorized access to user accounts
* Enable data manipulation or system compromise
* Create a security risk for our platform or users
## Ready to report?
Include detailed reproduction steps and explain the potential impact when submitting. This helps us understand and
address legitimate security concerns quickly.
================================================================================
URL: https://nextvisit.ai/help/business-associate-agreement
Title: Does Nextvisit provide a Business Associate Agreement (BAA)?
Description: How to request a Business Associate Agreement (BAA) for HIPAA-compliant use of Nextvisit AI.
Date: 2025-06-13
Topic: Security and data
================================================================================
Yes, Nextvisit will provide you with a Business Associate Agreement (BAA) if you are a subscribed user. It may take up
to 5 days to review and validate your request.
[Request BAA](https://app.commonpaper.com/pages/df056bba637366f3)
================================================================================
URL: https://nextvisit.ai/help/best-practices-recordings
Title: Best Practices for High-Quality Session Recordings
Description: Mic placement, session length, what to say aloud, and how to correct medication names after the visit.
Date: 2025-05-06
Topic: Getting started
================================================================================
## Set up your audio
- Use a microphone positioned close enough to clearly capture both you and your patient.
- Avoid background noise when possible. Quiet environments produce better results.
## Record the right length
- Sessions can be as short as 30 seconds or up to an hour.
- Record the full clinical conversation, especially the parts that matter for documentation.
## Include key clinical details
- Say the patient's name, presenting concern, diagnosis or condition, treatment plan, and any medications if you haven't
already set up a patient profile or context.
- Include any relevant follow-up instructions during the conversation.
## Medication names
- The AI may misinterpret drug names in the raw transcript.
- You can review and correct these post-session in your timeline or notes view.
The system filters small talk and background interruptions. Focus on the clinical conversation.
================================================================================
URL: https://nextvisit.ai/help/patient-consent
Title: Implementing Patient Consent
Description: A starting point for how you can inform your patients about Nextvisit AI with respect to patient consent.
Date: 2025-05-06
Topic: Getting started
================================================================================
## 1. Review your current consent forms
Locate the section(s) in your existing consent forms that cover data usage or patient privacy. Decide whether the
consent statement will be a new standalone section or folded into an existing one.
## 2. Customize the consent statement
Below is a patient-friendly statement you can adapt for your forms:
_I understand that my provider uses Nextvisit AI, a secure tool that helps document my care by creating clinical notes
from our conversation. This allows my provider to focus more on me during my visit. I consent to the audio recording of
my visit solely for this purpose. I know that my privacy is protected, and this recording is only used to improve my
care._
## 3. Tailor to your practice
**Updating your forms**
* Keep language simple, at approximately a 6th-grade reading level.
* Have it reviewed by legal professionals to confirm compliance with HIPAA, applicable state laws, and any other
relevant regulations.
* Add any additional notices or disclaimers required by law.
* Insert the statement into the predetermined section and match the formatting of your existing document.
**Educating your staff**
* Brief your team on the updated forms and the purpose of the consent statement so they can address patient questions
confidently.
* Provide staff with copies of the Patient FAQs Handout to use during patient interactions.
**Rolling out and monitoring**
* Begin using the updated forms with all new and returning patients.
* Collect staff feedback on any patient concerns so you can refine the process.
---
DISCLAIMER: This guide and the accompanying templates are provided by Nextvisit, Inc. for informational purposes only
and do not constitute legal advice. Providers are responsible for ensuring that any modifications or use of these
materials comply with all relevant laws and regulations, including but not limited to patient consent requirements and
privacy laws such as HIPAA and applicable state regulations. Nextvisit, Inc. assumes no liability for the use of this
guide or the templates.
================================================================================
URL: https://nextvisit.ai/help/recommended-devices
Title: Recommended Devices and Software
Description: New to Nextvisit? Make sure your device works flawlessly with us.
Date: 2025-05-05
Topic: Getting started
================================================================================
## Microphone requirements
Audio quality has a direct impact on transcription accuracy. When selecting a microphone:
* Use a microphone that can clearly pick up conversation with the patient.
* Prefer a noise-canceling microphone to reduce background noise.
* Test your microphone before starting a session.
## Recommended browsers
Use Google Chrome or Mozilla Firefox for best compatibility and performance. Both have been tested and work well with
Nextvisit AI.
================================================================================
URL: https://nextvisit.ai/help/data-protection
Title: How does Nextvisit AI protect patient data?
Description: Encryption, security protocols, and privacy controls that protect patient information in Nextvisit AI.
Date: 2025-04-29
Topic: Security and data
================================================================================
Nextvisit AI encrypts patient data at rest with AES-256 and in transit with TLS 1.2 or later.
================================================================================
URL: https://nextvisit.ai/help/hipaa-compliance
Title: Is Nextvisit AI HIPAA compliant?
Description: Yes. How Nextvisit AI meets HIPAA privacy and security requirements for patient information.
Date: 2025-04-29
Topic: Security and data
================================================================================
Yes, Nextvisit AI is HIPAA compliant. We follow all applicable HIPAA privacy and security requirements for protecting
patient information.
# News
================================================================================
URL: https://nextvisit.ai/news/the-checkup-august-27-2026
Title: The Checkup - August 27th, 2026 (v2.334.0)
Description: Nextvisit's August 27, 2026 release ages unsigned notes, adds Last time and a cross-titration advisor, plus Talk to Aria and a telehealth beta request.
Date: 2026-08-27
Type: Product update
================================================================================
The dashboard now tells you which unsigned notes are getting old. Pick a patient on Focus and you get a Last time
card before you hit Record. Medications can propose a day-by-day switch plan. Aria can talk, and telehealth has a
banner if your clinic is not on it yet.
## Unsigned notes, by age
You already knew how many notes still needed a signature. You could not see which of those were about to miss the
window. The pending list was newest first, so a note from Tuesday sank under this morning's drafts.
Focus now buckets unsigned, finished notes as fresh (under 24 hours), aging (24 to 47), or overdue (48 and up).
Overdue rows sort to the front. A chip opens Sessions filtered to overdue. The sessions list can badge aging and
overdue on the row.
Those windows are common clinic targets, not a claim about your payer's rule.
## Last time, before you walk in
Scope a patient on Focus and a Last time card fills from the chart: last completed note, current meds, active
allergies, and (when it is still fresh) homework or plan items from that visit. It does not call a model, so it does
not spend credits or make you wait.
If the last completed visit is 45 days old or older, the recap stays and the homework does not. The card will say
it has been a while. Treatment Pulse on the card follows the same reuse rule as the chart and is omitted when it is
stale. First session is a clean slate plus whatever is already on the meds and allergy lists.
## Switch an agent with a written taper
On the medications chart, Switch proposes a day-indexed cross-taper or washout for catalog antidepressants and the
common behavioral-health antipsychotics. It states the reason in plain language. You accept, modify structured
fields (strategy, washout, overlap, steps, interval, starting dose, pace), or reject.
Accept or modify writes the outgoing taper and the incoming titration onto the chart and schedules day-30 and
day-90 outcome prompts. Reject logs the decision and does not create those dues. MAOI pairs stay on a washout even
if you ask for a cross-taper.
This is a rules catalog, not a model. Unknown pairs are refused. Mood stabilizers, benzos, and stimulants are not
in this version. You still own the prescription.
## Talk to Aria
Aria chat has a Talk to Aria control on the full page and the floating widget. It is a live voice conversation, not
a dictation box that plays an MP3 after the fact. Chart lookup tools work over voice. Create, edit, and sign still
stay on typed chat.
The control is marked Beta. Playback is less choppy than the first cut this week, and a spoken turn stays one
bubble instead of stacking every pause as a new message.
## Aria note edits stay on the page
"Make edits with Aria" on an unsigned note used to hide the document and show a diff that looked like a code
review. The composer is now a quiet row that expands when you focus it. The note stays visible. You see a preview
only after Aria actually has edits. Apply, discard, and a short changes view sit on that same row. Escape collapses
it.
## Telehealth beta, if you want in
Built-in telehealth is still flagged off for most workspaces. If your clinic does not have it, a dismissible banner
lets you request access. That writes a signup and emails us. If you already have the flag, Telehealth shows under
Apps.
## Improvements
- Clinic owners can archive a former staff member from Settings → Workspace → Members. The person stays on the
roster as archived, loses clinic permissions, and their notes stay attributed to them. Restore puts them back as
staff. Remove is still the wrong tool for someone who left.
- Printed notes now put date of birth on the letterhead and repeat name and DOB on every page.
## Bug fixes
- The sessions date filter was comparing UTC storage to your calendar day, so an evening visit in the US could
vanish. It now uses your timezone.
- Initial-evaluation notes with MSE turned on in preferences were skipping the exam when the template had no
place for the results. The exam runs and the results land in the note.
- A newly created custom template could leave the Custom tab empty until a refresh. The list updates in place.
## Availability
Unsigned-note aging, Last time, agent switch, archive, and the Aria note-edit composer are in this week's build.
Talk to Aria is beta and needs voice enabled on the workspace. Telehealth remains request-only unless your team
flag is already on. The this-visit prescribe pad from last week is still rolling out on its own flag.
Questions or feedback on anything in this release? Tell us through the in-app feedback button. It goes to the people
who built this.
================================================================================
URL: https://nextvisit.ai/news/the-checkup-august-21-2026
Title: The Checkup - August 21st, 2026 (v2.325.0)
Description: Nextvisit's August 21, 2026 release rebuilds prescribe as a this-visit pad, imports a record during patient create, and redesigns sign-in.
Date: 2026-08-21
Type: Product update
================================================================================
Psychiatrists get a this-visit Rx pad instead of a second copy of the med list. Creating a patient can start from
a PDF or pasted record. Sign-in is a new shell. A few list and credit nits are at the end.
## Prescribe is a this-visit pad
The first prescribe surface glued the chart table to order entry. That made the list the work, and the thing you
were trying to send sat in a leftover rail.
Prescribe is now a pad for what leaves this visit. The chart meds stay in a review strip. Renew on a maintenance
med drops a line onto the pad. A C-II still opens compose and the EPCS path. The visit-end sheet uses the same pad,
so you are not restaging the same three lines in two places.
From a session, the Medications widget has Prescribe. On the chart, open the patient and use Prescribe there.
## Import a record when you create a patient
The New Patient dialog can take a PDF or pasted chart text and prefill name, date of birth, administrative sex,
notes, medications, allergies, and insurance. You can edit every field before anything is saved. No patient is
created until you submit the form you already know.
The import runs in the background with a live status. If you finish the create, the PDF lands in that patient's
Documents. Abandoned imports drop after a day.
## Sign-in was rebuilt
Sign-in shares one dark shell now, so the first page matches the rest of the product. Google and Apple are still
on that screen. It is the same account.
## Improvements
- A few generation tasks moved to newer models. You do not pick this; drafts that used those models just use the
newer ones.
- If a post-visit summary cannot run because the workspace is out of credits, the notice says that plainly instead
of failing in a generic way.
- Searching patients or sessions no longer kills infinite scroll. The list keeps its scroller when filters reset,
so you do not fall back to a short page and think that is the whole roster.
## Bug fixes
- Aria chat on a team conversation could not stream if it lost that team's context. Streams stay pinned to the
chat's team.
- An Aria widget reply that never left the "thinking" state could vanish. The answer now persists.
- Deleting a user no longer blanks the provider name on their old encounters.
- Expanding a patient or encounter macro, and listing healthcare context, stay inside the current team.
## Availability
The this-visit prescribe pad is rolling out behind the eRx UI flag, so some clinics will still see the older
composer. Patient-record import is in the New Patient dialog in this week's build. The new sign-in is on for
everyone.
Questions or feedback on anything in this release? Tell us through the in-app feedback button. It goes to the people
who built this.
================================================================================
URL: https://nextvisit.ai/news/the-checkup-july-17-2026
Title: The Checkup - July 17th, 2026 (v2.321.2)
Description: Nextvisit's July 17, 2026 release opens Duplicate Patients, adds in-call telehealth extras and file sharing, and shows Aria chat as pending while it replies.
Date: 2026-07-17
Type: Product update
================================================================================
Duplicate Patients is in the sidebar this week, and the matcher is less likely to pair the wrong Mary or miss a
Sam. Telehealth picked up the in-call pieces people actually use on a video visit, and Files can share a link
instead of downloading a copy. Aria chat also stopped looking idle the moment you send.
## Duplicate Patients is in the sidebar
Every practice accumulates a second chart. Intake spelled the last name differently. Someone created "Bob" next to
"Robert." One history is then split across two records.
Duplicate Patients scans the roster, shows why it thinks two charts are the same person, and lets you merge them.
Encounters, documents, telehealth links, and healthcare context move onto the chart you keep, with an audit trail.
You can archive a leftover instead of merging it.
Matching is tighter than the first pass. Sammy and Samuel can match again; a nickname-map bug had blocked that.
Mary Ann and Mary Beth no longer collapse to a medium-confidence "Mary." A garbage date of birth no longer knocks a
perfect name match down. Failed scans say they failed, with a retry, instead of claiming there are no duplicates.
## Telehealth: chat, files, blur, and attestation of consent to record
If your workspace has built-in telehealth, the call itself has more than video now. In-call chat, peer file share,
background blur, and a provider photo capture are on the visit. Patients can get a web push when you are ready.
Recording, encounter linking, and photo capture wait on a consent attestation that the server stores. Rejoining a
session does not skip that step. If the connection drops to audio-only, a banner says so instead of leaving you to
guess why the picture froze.
The patient buttons are larger, and a screen reader can follow chat and file events during the visit.
## Files can be shared
Files grew a details panel, empty states, and keyboard shortcuts. You can share a file, restore one from trash, and
open a shared link.
## Improvements
- Notes and Aria tasks moved to newer models this week. You do not change a setting. Drafts that used those models
just use the newer ones.
- Prompt versions in the template editor are append-only now, with restore, so an edit you regret is not gone.
- Aria chat shows a pending state the moment you send, instead of sitting quiet until the first token arrives.
## Bug fixes
- Duplicate scan group ids stay stable across rescans, so a merge button cannot read "Merge 0" after the list
refreshes.
## Availability
Duplicate Patients appears when the workspace flag is on. Built-in telehealth extras only show if your clinic
already has telehealth. File sharing is in this week's build for workspaces that use Files.
Questions or feedback on anything in this release? Tell us through the in-app feedback button. It goes to the people
who built this.
================================================================================
URL: https://nextvisit.ai/news/the-checkup-july-10-2026
Title: The Checkup - July 10th, 2026 (v2.317.1)
Description: Nextvisit's July 10, 2026 release adds a patient post-visit summary, a medication timeline, faster picking, and the first built-in telehealth build.
Date: 2026-07-10
Type: Product update
================================================================================
There is a post-visit summary you can generate, edit, and print, a medication timeline that shows dose and gaps
instead of a flat list, and a patient list that finally filters. Built-in telehealth also exists as a real visit
now, still gated per workspace.
## Post-visit summary: something the patient can leave with
The note you sign is for the chart. The thing a patient asks for in the hallway is shorter: what you covered, what
changed, when to come back.
You can generate a post-visit summary from the encounter, open it as a document or a dialog, edit the next
appointment on it, and print a PDF that includes that date. Regenerating the summary keeps the appointment you
already set, so a reprint does not wipe the follow-up you typed.
## Medication timeline
The medications page can show a timeline, not just the current list. Dose is drawn by strength, so a titration is
visible as a thicker or thinner run. Periods off a med show as gaps instead of looking like the row ended. It loads
in stages so a ten-year chart does not stall the page, and you can refresh it after you edit.
## Finding the next patient, and starting the next session, got faster
The patient list has a filter bar: search, sort, and facet filters, including tag color. That is the difference
between scrolling for "the Tuesday TMS people" and actually filtering to them.
On the recorder, the template grid supports quick start and keyboard navigation. Session info now shows the template
you picked, with badges and hotkeys, so you can confirm the note type without backing out of the setup screen.
## Built-in telehealth, first cut
You can run a video visit inside Nextvisit instead of opening Zoom. The patient gets a link, lands in a green room
where they can check camera, mic, and speaker, and waits until you admit them. You get a queue, a link manager, room
settings, and in-call controls next to the recorder. Stereo capture is there so both sides can transcribe cleanly.
This is the first build. It is behind a workspace flag. If you do not see it, that is expected.
## Improvements
- The login screen cycles a few short lines about notes. It respects reduced motion and pauses when the tab is
hidden.
- Focus dashboard widgets can show pending encounter counts and the template on those rows.
- Navigation is faster on the pages we have hardened, including the session view.
## Bug fixes
- The patient list on a phone was drawing the avatar twice. One face per row now.
- Settings, usage, and a few other pages were rendering empty for a beat after the client upgrade. They show a
skeleton until the data is there.
## Availability
Post-visit summaries, the medication timeline, patient filters, and the recorder template work are in this week's
build. Built-in telehealth is rolling out per workspace. If your sidebar does not have it yet, it is not on for your
clinic.
Questions or feedback on anything in this release? Tell us through the in-app feedback button. It goes to the people
who built this.
================================================================================
URL: https://nextvisit.ai/news/the-checkup-june-19-2026
Title: The Checkup - June 19th, 2026 (v2.240.2)
Description: Nextvisit's June 19, 2026 release adds a chart-proposal inbox, a rearrangeable dashboard, EHR sync review, and a heading outline for long notes.
Date: 2026-06-19
Type: Product update
================================================================================
Encounter extraction can now propose med, condition, and allergy changes for you to accept or reject. The dashboard
can be rearranged, or swapped for a focus view built around the next session. Practices pulling charts from an EHR
get a review inbox instead of a silent overwrite.
## Chart proposals: the visit can update the chart without you retyping it
A patient mentions they stopped the sertraline two weeks ago, or that they started a prazosin from another clinic. You
used to finish the note and then go hunt the medications list. Now the encounter can put those changes in a proposal
inbox on the chart: add, edit, or discontinue, with the rationale attached.
Accept one, reject one, or take the lot. Duplicate creates from the same run get collapsed. A later encounter that
proposes the same add supersedes the earlier one, so you are not accepting the same med twice. Pending proposals stay
visible until you deal with them, including a count on the patient.
Auto-accept is on by default so a clean extraction can land without a click. Turn it off in Settings if you want every
write to wait for you. You still decide what lands in the record.
## A dashboard you can actually set up
Home is no longer one fixed layout. Classic mode has widgets you can rearrange, show, or hide, including a clock and
a greeting that follows the time of day. Focus mode is the other option: recent sessions, a session launcher, and a
patient scope search when you already know who is next.
There is a "try the new dashboard" control if you want to switch without hunting settings. If you liked the old
arrangement, it is still there.
## EHR sync has a review inbox
Practices that pull charts from an outside EHR now get a compiled roster and a per-patient chart with a section rail
and a diff view. Matches and conflicts land in one inbox. You can pin a field, preview a sync plan, approve in bulk,
or roll a write back.
It is meant for the person who has to decide whether the EHR's allergy list or yours is right, not for a silent merge
at 2 a.m.
## Long notes got an outline
Open a long note or a template and a heading rail tracks where you are. Click a heading to jump. Tables of contents
in the editor follow the same headings, so a twelve-section intake is not one long scroll.
## Improvements
- Patient chart pages (demographics, medications, conditions, allergies, documents, scales) keep a sticky header as
you scroll, and the collapsed patient nav can open a flyout instead of making you expand the rail.
- Demographics uses segmented tabs with a sliding pill, which is less fiddly on a tablet.
- Sessions has a filter bar, including capture source, so finding last Tuesday's Zoom visit is a filter instead of a
hunt.
- Support moved into the sidebar. The floating launcher is gone. If the messenger is down, a fallback contact form
still reaches us.
- Display settings let you pick how the patient nav is laid out.
- You can put an encounter in input review and keep the note as a draft until that review is done. The encounters
table shows pending-review and awaiting-review states.
## Bug fixes
- Chart proposals no longer stack the same create twice in one run, and accepting a later duplicate retires the
earlier one.
- The medication form no longer submits on Enter when you are still in a field.
- Team switcher on smaller screens was clipping its menu. It stays on screen now.
- The patient profile had a broken markup tag that could scramble the layout. Closed.
- Timeline queries that sent a non-integer encounter id no longer fail the page.
## Availability
Chart proposals, the new dashboard layouts, and EHR sync review are in this week's build. Auto-accept for proposals
is on unless you change it. EHR sync only appears if your workspace is set up to pull from an outside chart.
Questions or feedback on anything in this release? Tell us through the in-app feedback button. It goes to the people
who built this.
================================================================================
URL: https://nextvisit.ai/news/the-checkup-june-5-2026
Title: The Checkup - June 5th, 2026 (v2.237.2)
Description: Nextvisit's June 5, 2026 release adds Aria Highlights, Note-Ready, and Golden Thread to the live session, and rebuilds medications, conditions, and Aria chat.
Date: 2026-06-05
Type: Product update
================================================================================
There are three new Aria surfaces for the visit itself, rebuilt medications, conditions, and timeline views in the
chart, and an Aria chat that works like a workspace. Improvements and fixes are at the end.
## Aria Highlights: the moments you would have stopped to write down
Aria Highlights collects those moments as they're said. It surfaces short, categorized cards: reason for visit, a
reported symptom, a stressor, an impact on daily functioning, a strength or support, a medication, a stated plan. When
a moment has a clean verbatim line, the card keeps the exact wording instead of a paraphrase. Pin a card and it carries
through to the note Nextvisit generates. Dismiss it with one tap.
On a routine visit it's a count in the corner. On a complicated one it's the list you scan before you close the note.
It reports what was said and nothing more: it does not score severity, flag risk, or suggest a diagnosis. You decide
what belongs in the record.
## Note-Ready: know your note is covered before the session ends
Next to Highlights, a progress ring fills in as the conversation covers what your selected note type needs. A glance
tells you whether your documentation is on track, or whether a gap is worth one more question before the visit ends
rather than an addendum tomorrow.
## Golden Thread: check your note's defensibility before you sign
When payers audit behavioral health notes, they're checking one thing: does the assessment connect to the plan, the
plan to the intervention, and the intervention to documented progress and medical necessity? Golden Thread reads your
finished note from the session view and shows where that through-line holds and where it thins, while the session is
still fresh and the note is still unsigned. Run it when you want it, dismiss it when you don't.
## Aria chat is now a full workspace
Aria chat now remembers. Every conversation is saved and searchable, and picks up where you left off.
You can attach files and images directly to a conversation, and everything Aria produces lives in a new artifacts
gallery with previews, so Tuesday's discharge summary draft is one click away instead of buried in scrollback. You can
edit and resend a message in place if you catch a typo, or regenerate a response if you want a different take. A
command palette puts actions on keyboard shortcuts, and a new appearance menu lets you pick themes and a reading font,
including two serif options we host ourselves so no outside font service ever sees your traffic.
Aria also moved to our newest model and can now ground its answers with web search where that helps.
## Medications, conditions, and the timeline got rebuilt
The parts of the chart you touch every day were rebuilt:
- Medications has filters and sorting, a redesigned add/edit form with sig and quantity fields, and quick fill from the
patient's recent prescriptions, so re-documenting a refill takes seconds instead of retyping the whole entry. You can
search prescription history from the form and delete entries you no longer need.
- Conditions now track treatment status and diagnosis date, support editing in place, and filter and sort the same way
medications do.
- Dates accept how people actually talk. Type "3 weeks ago" or "last March" and the field works it out.
- The patient timeline was redesigned with better filtering and navigation, so finding that one encounter from last
fall no longer means scrolling through everything since.
- The allergies and notes dialogs were upgraded to match.
## See exactly where your credits go
Settings now has a Usage page: the current period, what's been used and what remains, a breakdown by activity, a usage
trend, and the full transaction history. A small indicator keeps the running total visible, so the end of a billing
cycle is never a surprise.
## Coming soon: duplicate patient cleanup
We've built a Duplicate Patients tool that finds likely matches, shows why it thinks they match and how confident it
is, and merges them into one chart, moving every encounter, document, and record with an audit trail. It's in final
testing and will start appearing in the sidebar over the coming weeks.
## Improvements
- The login and signup screens were redesigned: cleaner layout, single sign-on options up front, and a lighter password
strength meter.
- Notes and other AI tasks moved to newer models tuned per task, so generation quality improves without you changing
anything.
- Timeline content is now sanitized before rendering, an extra layer of protection in the chart.
- Search indexing retries automatically on transient failures, so new records show up in search more reliably.
## Bug fixes
- Requesting a password reset showed the success message styled as a red error. It now reads as the confirmation it is.
- Aria chat could occasionally drop a response mid-thought. Fixed.
- The model selector in Aria chat sometimes needed two clicks to register. One click now works.
- Phase details in the patient summary could render incorrectly. They display properly now.
## Availability
Aria Highlights, Note-Ready, and Golden Thread are rolling out gradually, so if you don't see them yet, they're on
their way. Where they're available they work with no setup, and each one can be switched off at any time.
Questions or feedback on anything in this release? Tell us through the in-app feedback button. It goes to the people
who built this.
================================================================================
URL: https://nextvisit.ai/news/introducing-aria-highlights
Title: Introducing Aria Highlights
Description: A live capture feed in the Nextvisit recorder that collects the moments a behavioral health note relies on, as they are said.
Date: 2026-06-01
Type: Product update
================================================================================
Aria Highlights is a live capture feed in the Nextvisit recorder. It listens with the provider and collects the moments a behavioral health note tends to rely on, as they are said: the sentence that explains why they came, the med they stopped two weeks ago, the thing at home that pulled the week apart.
## What it does
As the conversation moves, Aria Highlights surfaces short, categorized cards: the reason for the visit, a reported symptom, a stressor, an impact on daily functioning, a strength or support, a medication, a stated plan. When a moment has a clean verbatim line, the card keeps that wording instead of a paraphrase.
Each card has one primary action: pin it to the note. Pinned moments drop into the session record and carry through to the documentation Nextvisit generates, so a dose, a date, a name, or a commitment made out loud lands the way it was said. Anything the provider does not want is one tap to dismiss.
The feed asks for nothing. No sound, no alert, no modal. It sits in the corner of the recorder next to Aria Note-Ready, which shows how well the session is covering what the selected note type needs.
## Stays out of the way
Aria Highlights shows a quiet count while the provider works and expands into the full feed only when they reach for it, at a pause or as the session ends. It can be turned off for a session at any time. On a routine visit it is a number in the corner. On a complex one it is the list you scan before you close.
## A documentation mirror
Aria Highlights reports what was said. It does not assess clinical adequacy, rate severity, score risk, decide a diagnosis, or judge whether a note is complete or compliant, and it makes no claim about reimbursement. The categories are documentation categories. The provider reads, edits, and decides what belongs in the record.
## Availability
Aria Highlights is rolling out now to behavioral health providers on the Nextvisit live recorder and will expand to the full base over the coming weeks. Where it is available it is on by default, and it can be switched off for any session. No setup is required.
================================================================================
URL: https://nextvisit.ai/news/nextvisit-ai-launches-ios-app-putting-clinical-documentation-in-providers-pockets
Title: Nextvisit AI Launches iOS App, Putting Clinical Documentation in Providers' Pockets
Description: Nextvisit launched an iOS app, ClinicalConcert, so providers can record a visit and get a structured note from the phone.
Date: 2025-07-04
Type: Product update
================================================================================
## iOS app lets providers record and transcribe patient notes from anywhere
NEW YORK - July 3, 2025 - Nextvisit launched an iOS app that brings clinical documentation to the iPhone. Providers can record a conversation and get a formatted note in seconds, in the office, on a home visit, or between sites.
The app connects to the Nextvisit web platform. Clinicians speak during or after the encounter, and the AI generates a structured note.
"We've put Nextvisit AI in every provider's pocket," says Ryan Yannelli, Co-founder & CTO of Nextvisit AI. "Whether they're conducting home visits, working in multiple locations, or just need to capture notes right after a session, documentation doesn't have to wait."
Download [ClinicalConcert](https://apps.apple.com/us/app/clinicalconcert/id6743886877) from the App Store. The platform is free to start at , with no credit card required.
**About Nextvisit AI**
Founded by psychiatrist Dr. Faisal Rafiq and software engineer Ryan Yannelli, Nextvisit turns doctor-patient conversations into clinical notes.
================================================================================
URL: https://nextvisit.ai/news/announcing-our-free-plan
Title: Announcing Our Free Plan
Description: Nextvisit is free for individual clinicians: 15 SOAP notes a month, basic analytics, no credit card.
Date: 2025-06-13
Type: Announcement
================================================================================
Nextvisit is now free for individual clinicians, no credit card required.
The free tier includes up to 15 SOAP notes per month, basic AI analytics, and single-provider setup. Use it on real sessions before you pick a paid plan.
[See what's included](https://nextvisit.ai/pricing)
================================================================================
URL: https://nextvisit.ai/news/nextvisit-ai-teases-major-ui-refresh-to-streamline-clinical-workflows
Title: Nextvisit AI Teases Major UI Refresh to Streamline Clinical Workflows
Description: Nextvisit is preparing a UI refresh with a cleaner layout and fewer clicks on common documentation tasks. Rollout is in the coming weeks.
Date: 2025-05-18
Type: Product update
================================================================================
Nextvisit is preparing a new interface for documentation and care coordination: a cleaner layout, faster navigation, and workflows built around how providers already work.
"We've listened closely to our users, and this UI refresh is all about making their day-to-day work simpler and more intuitive," said Ryan Yannelli, CTO and cofounder of Nextvisit AI. "The new design gives clinicians a clear path to complete their documentation, manage patient information, and collaborate with teammates, without the friction that comes with legacy software."
The updated UI will roll out in the coming weeks. The design focuses on navigation for common documentation tasks. More details will follow.
For more information, visit [nextvisit.ai](https://nextvisit.ai/) or
contact [press@nextvisit.ai](mailto:press@nextvisit.ai).
================================================================================
URL: https://nextvisit.ai/news/nextvisit-ai-helps-addiction-providers-deliver-better-care-not-just-paperwork
Title: Nextvisit AI Releases MAT Note for Addiction Providers
Description: Nextvisit released a MAT Note that puts medication compliance, toxicology results, counseling progress, and regulatory requirements in one view.
Date: 2025-05-09
Type: Press release
================================================================================
## New MAT Note supports review of medication compliance, toxicology results, and counseling progress
NEW YORK - May 9, 2025 - Nextvisit released a MAT Note that puts medication compliance, toxicology results, and counseling progress in one structured view, so addiction providers can see the pattern before they write the next dose.
"This is about more than saving time. It's about seeing the complete patient story," says Ryan Yannelli, Co-founder & CTO of Nextvisit AI. "When providers can quickly spot patterns across medication compliance, toxicology results, and counseling progress, they make more informed decisions that directly impact recovery outcomes."
The structured MAT Note lets providers review regulatory requirements in one place.
Request a demonstration of the MAT Note.
**About Nextvisit AI:** Founded by psychiatrist Dr. Faisal Rafiq and software engineer Ryan Yannelli, Nextvisit turns doctor-patient conversations into clinical notes.
================================================================================
URL: https://nextvisit.ai/news/nextvisit-ai-joins-intercom-startup-program-accelerates-support-capabilities-to-meet-growing-demand-in-behavioral-health-tech
Title: Nextvisit AI Launches Intercom Help Center and Live Chat
Description: Nextvisit joined Intercom's Early Stage Startup Program and launched a help center, live chat, product updates, and Fin AI ticket resolution.
Date: 2025-05-08
Type: Press release
================================================================================
_New York, NY - May 7, 2025_ - Nextvisit Inc. has been accepted into Intercom's Early Stage Startup Program. The program scales support as the user base grows among therapists, counselors, and other non-prescribing clinicians.
Within a few days of joining, Nextvisit launched a help center, live chat, a product updates feed, and automated ticket resolution using Intercom's Fin AI. Providers can get answers without waiting on the team.
"This came at the perfect time," said **Faisal Rafiq**, CEO and Co-founder of Nextvisit AI. "Our user base is expanding fast, and Intercom lets us deliver solid support without slowing down the pace of development."
Nextvisit's AI scribe generates SOAP notes from patient conversations.
"We are building for real-world problems," said **Ryan Yannelli**, CTO and Co-founder. "Intercom lets us handle support at scale while staying focused on what providers actually need from the product."
To learn more, visit