Individual Therapy Billing: CPT 90837, 90834, and 90832 Explained
Mariam Shaker6 min read
Short answer: CPT 90837 is individual psychotherapy lasting 53 minutes or more, 90834 covers 38 to 52 minutes, and 90832 covers 16 to 37 minutes. The code is decided by actual session time, not by the appointment slot on the calendar, and 90837 draws more payer scrutiny than the other two because it reimburses the most and audits the documented minutes most closely.
Some of what follows comes from the codes themselves. The rest comes from somewhere Oasys is lucky to sit: Oasys runs at the infrastructure layer of working practices, and we talk constantly with therapists and practice owners about exactly this kind of billing mechanic, the ones that look simple until a claim gets denied or a payer opens a documentation request.
The three codes
CPT 90837, Individual psychotherapy, 60 minutes
53 minutes or more, face-to-face with the patient (or via telehealth where permitted). This is the highest-reimbursing common individual therapy code, which is exactly why it gets audited more than 90834 or 90832. A payer that sees a practice billing 90837 for every single session, every client, every week, is looking at a pattern that invites a documentation request.
CPT 90834, Individual psychotherapy, 45 minutes
38 to 52 minutes. The most commonly billed individual therapy code industry-wide, and the one most clinicians' natural session length actually falls into.
CPT 90832, Individual psychotherapy, 30 minutes
16 to 37 minutes. Used less often in outpatient therapy, more common for brief check-ins, medication-adjacent visits, or specific payer requirements.
All three are claim types Oasys generates directly from the signed note, so which one applies follows the same documented session length a payer would check against.
What decides the code: minutes, not the calendar slot
A session scheduled for 60 minutes that actually runs 48 minutes of clinical content bills as 90834, not 90837. The code follows the documented time, and the documentation has to say what that time actually was, not just repeat the code's own name back as the note.
The note needs an actual start and end time, or a total minutes figure, that a payer can check against the code billed. "60-minute individual session" typed at the top of every note regardless of what happened in the room is exactly the pattern payers flag, because it makes the same claim every time rather than a variable one.
Other platforms may be built differently, and leave the CPT code as a field billing staff set by hand after the fact. Oasys captures session time as part of the note itself, so the minutes behind a code are the minutes a clinician actually documented, not a separate number entered downstream.
Why 90837 gets audited more than the others
Three things converge on 90837 specifically: it pays more per session than 90834 or 90832, a practice that bills it near-universally has an unusually flat distribution across an inherently variable thing (real session length), and CMS and commercial payers have both published guidance flagging exactly this pattern as an audit trigger. None of that means 90837 is wrong to bill. It means the documentation has to earn it every time, not by default.
Other platforms may be built differently, but Oasys ties the CPT code on the claim to the same note the clinician actually signed, rather than a separate billing-side field a staff member sets independently. If the note does not support 53 minutes, the mismatch is visible before the claim goes out, not after a payer flags it.
Add-on and adjacent codes worth knowing
90833 and 90836 are the psychotherapy add-on codes used alongside an evaluation and management (E/M) visit, typically in psychiatric or medication-management settings where a prescriber also does therapy in the same visit. They follow the same time bands as 90832 and 90834 respectively, but are billed in addition to the E/M code, not instead of it. Most outpatient therapy practices without prescribing clinicians will not use these; group practices with psychiatric providers should confirm with each payer which combinations are reimbursed.
Common billing mistakes with individual codes
Billing 90837 by default regardless of actual session length. The code should follow the clock, not a practice-wide habit.
Documentation that states a code's name instead of the session's actual content and duration. "90837, individual psychotherapy" is not documentation. What happened in 53-plus minutes is.
Not adjusting the code when a session runs short. A client who leaves at 40 minutes changes the code to 90834, not a shorter version of 90837.
Confusing time-based codes with add-on codes. 90833 and 90836 are billed with an E/M visit, not as standalone individual therapy codes.
These mistakes share a root cause: the CPT code living apart from the documentation that should justify it. Oasys keeps the two in one record specifically so a mismatch is something a clinician can catch before signing, not something a payer catches months later.
What to ask before you commit to a platform
Does the system pull the CPT code from the same session record the clinician documents, or does billing staff set it separately from what the note says?
Can a clinician see, at the point of signing a note, whether the documented time supports the code being billed?
Does the platform distinguish 90832, 90834, and 90837 automatically, or does someone have to remember the time bands and select manually every session?
If a payer opens a documentation request, can the practice pull the note and the claim together, or are they in two disconnected systems?
Oasys answers the first and last of those by design: the claim is generated from the signed note, so pulling one pulls the other.
How Oasys handles it
Oasys generates the CPT-coded claim directly from the signed note rather than from a separate billing-side entry, so the code billed and the documentation behind it are the same record, not two versions that can quietly drift apart. Pricing is $80 per clinician per month on an annual contract, with no per-claim fee and no add-on tier for insurance billing.
FAQ
What is the difference between CPT 90834 and 90837? Time. 90834 covers 38 to 52 minutes of individual psychotherapy; 90837 covers 53 minutes or more. The clinical content can be similar; the code follows the documented duration.
Why does 90837 get audited more than 90834? It reimburses more per session, and a practice that bills it for nearly every session regardless of actual length produces an unusually uniform pattern for something that should vary. CMS and commercial payers have both flagged this specific pattern in audit guidance.
Can I bill 90837 for a 50-minute session? No. 90837 requires 53 minutes or more of documented time. A 50-minute session bills as 90834.
What is CPT 90832 used for? Individual psychotherapy lasting 16 to 37 minutes, typically a brief check-in or a shorter session than standard outpatient therapy, and it is billed less often than 90834 or 90837 in general outpatient practice.
Do I need to document the exact session length for every code? Yes. The safest practice is documenting an actual start and end time or a total minutes figure for every session, regardless of which code is billed, so the note itself supports the code if a payer ever asks.
What does insurance billing cost on a platform like Oasys? Oasys prices at $80 per clinician per month on an annual contract, with billing included and no per-claim fee, so correcting a miscoded session does not carry a separate transaction cost.
The code is not a description of the appointment type. It is a claim about what actually happened in the room, and the documentation is the only thing that can back that claim up later.
Related articles

Best Billing Software for Behavioral Health Practices (2026)
The best billing software for a behavioral health practice is judged by what happens after a claim is submitted, not before. A guide to eligibility, denials, and real cost.

Should a Group Practice Buy a Standalone AI Scribe or an EHR With AI Built In?
A standalone scribe adds a second vendor touching every session. An EHR with AI built in keeps the note, the chart, and the claim in one system.

How Group Therapy Practices Reduce Insurance Claim Denials
Most denials are not random payer behavior. They trace back to a handful of preventable causes, and each one has a specific, checkable fix.

Best Practice Management Software for Therapists (2026)
The best practice management software for therapists depends on clinician count, not features on a marketing page. A guide to what actually separates the options at each stage.

Can AI Write Therapy Notes That Hold Up to an Insurance Audit?
An auditor does not check whether AI wrote a note. They check whether a clinician actually reviewed it, signed it, and whether it supports the billed code.

Does Your EHR Track Supervision Hours Toward Licensure?
No. Your EHR almost certainly does not accumulate supervision hours toward a licensing board, and Oasys does not either.

HIPAA Compliant Therapy Platform: What to Actually Verify (2026)
"HIPAA compliant" is not a certification a vendor earns once. It is a signed BAA plus specific, checkable safeguards. Here is what to actually verify before you sign.

Superbills for Therapists: What They Are and When to Use One
A superbill is an itemized receipt a client submits to their own insurer for reimbursement. Here is what has to be on one, and when a practice needs them at all.

Individual Therapy Billing: CPT 90837, 90834, and 90832 Explained
CPT 90837 covers individual psychotherapy of 53 minutes or more. Here is how it differs from 90834 and 90832, and why 90837 draws more payer scrutiny than either.

TheraNest vs SimplePractice: Which Is Better for Your Practice?
TheraNest prices on active clients rather than seats, which suits fluctuating caseloads. Plus a migration warning worth reading before you commit.

How IM+CANS batch export works, and why double entry isn't required
Getting a completed IM+CANS assessment into Illinois HFS's provider portal does not have to mean retyping the same client's data a second time.

Best EHR for IM+CANS (IMCANS) in Illinois (2026): what actually matters
The best IMCANS EHR is not the one with an IM+CANS-shaped form. It is the one that treats IM+CANS as five separate problems stacked together: the roughly 700-field instrument…

What IM+CANS (IMCANS) actually requires from an Illinois practice
IM+CANS is not optional paperwork. It is the single instrument the Illinois Department of Healthcare and Family Services (HFS) designated for Integrated Assessment and Treatment…

How to Grow a Group Therapy Practice From 5 to 25 Clinicians Without It Breaking
The practice does not break from headcount. It breaks from the workflows that were fine at 5 clinicians and were never redesigned for 15 or 25.

Credentialing vs. Payer Enrollment vs. Clearinghouse Enrollment: Why They're Not the Same
Credentialing, payer enrollment, and clearinghouse enrollment are three separate steps that happen in sequence, not one process with three names.

The Weekly KPIs a Group Therapy Practice Owner Should Actually Watch
A group therapy practice owner should watch four numbers every week: utilization, no-show rate, caseload per clinician, and billing health.

TheraPlatform vs SimplePractice: Which Suits Your Practice?
TheraPlatform was built around teletherapy rather than retrofitted for it. Whether that is worth it comes down to one number, your virtual share.

TheraNest Reviews: What Therapists Actually Say
Active-client pricing makes TheraNest cheaper for variable caseloads. Reviews are solid on function, softer on polish, and mixed on migration.

Sessions Health Reviews: What Therapists Actually Say
Reviewers consistently praise price, simplicity and support. The one consistent criticism is reporting, and whether that matters depends on your size.

Jane App Reviews: What Therapists Actually Say
Praise concentrates on scheduling and running a mixed clinic from one system. Criticism concentrates on mental health depth and treatment plans.

Sessions Health vs SimplePractice: A Practical Comparison
Sessions Health is the better value for small practices, with free admin seats and no feature gating. The trade-off is reporting depth.

Jane App vs SimplePractice: Which Fits a Therapy Practice?
Jane wins on scheduling and multi-disciplinary clinics. SimplePractice wins on therapy-specific depth. The treatment-plan gap decides most of it.

Oasys vs TherapyNotes: Which Is Better for a Group Practice?
TherapyNotes is genuinely well built. Here is where it stops fitting a growing group practice, and what the per-transaction fees really add up to.

Oasys vs SimplePractice: Which Is Better for a Group Practice?
A straight comparison for practices with five or more clinicians, covering supervision, permissions, payroll and what switching actually involves.

SimplePractice Reviews: What Practices Actually Say in 2026
SimplePractice is excellent for solo therapists and strains for group practices. What practices praise, what they criticise, and where the line falls.

SimplePractice vs TherapyNotes: An Honest Comparison for 2026
Two good products that break in the same place. An honest comparison of pricing, supervision and billing, with our own stake clearly labelled.

SimplePractice Alternatives for Group Practices
Why group practices outgrow SimplePractice, what the realistic alternatives are, and how to move eight years of records without breaking the practice.

What SimplePractice and TherapyNotes Actually Cost a Group Practice
The subscription is the smallest part of the bill. What group practices actually pay once per-claim fees, text charges and workarounds are counted.

Best EHR for Group Therapy Practices (2026)
Group practices break in predictable places as they grow. A ranked guide organised around what fails at five, fifteen and thirty clinicians.

Mental Health EHRs: The Complete Comparison Guide (2026)
There is no best mental health EHR, only a best one for your size. A guide organised by practice size, with five tests demos are not designed for.

Mental Health EHR Requirements: What to Actually Demand
The requirements that actually matter, grouped by what breaks at five, fifteen and thirty clinicians, plus a ten-item checklist to run live in a demo.

Mental Health Treatment Plan Examples
The four parts of a usable plan, worked examples for anxiety, depression, trauma and substance use, and the goal formula that makes progress notes write themselves.

Group Therapy Billing: CPT 90853, 90847, and 90846 Explained
CPT 90853, 90847 and 90846 explained, plus the five-minute test that shows whether your EHR can actually bill a group session without manual work.

DAP Notes: Format, Examples, and When to Use Them
Data, Assessment, Plan. A worked example, how DAP compares with SOAP and BIRP, and the mistakes that make notes fail during a payer audit.

SOAP Notes: Format, Examples, and How to Write Them Faster
The four sections, a worked example with a real assessment, and why notes take so long. Hint: it is almost never the template that slows you down.

How Tate Psychotherapy left SimplePractice without missing a session
How a solo NYC psychotherapy practice switched from SimplePractice with zero missed sessions, then put notes and billing on autopilot with Oasys.

How SoCal Soul launched a fully equipped practice from day one
How a two-clinician startup launched with Oasys' Managed RCM handling all billing, and wearable health data built into the client record from day one.

How Lowry Hill got an EMR shaped around how its clinicians actually work
How an eclectic group practice moved off TheraNest, kept billing in-house on Oasys, and got custom workflows, AI documentation, and fast support.

How Camino Nuevo cut documentation time by ~40% with Oasys
How a school-based mental health program cut documentation time ~40%, saved clinicians 8–9 hours a week, and fully onboarded in under one week.

Is it actually safe to use AI for therapy notes?
AI therapy notes can be HIPAA compliant, but safety depends on the tool, not the category. Here are three facts to verify with any vendor.

What actually goes wrong when therapy practices switch EHRs
EHR migrations fail before the data moves. Here's what actually breaks — consent gates, lost calendars, flattened notes — and how to clear it.

Own your brand vs. rent a network: the real economics
Therapy networks take 20–30% of billing revenue, per session, indefinitely. Here's how to decide when to rent a network and when to own your brand.

The group-practice scaling wall: what breaks at 5, 15, and 30 clinicians
Solo-built EHRs break at predictable thresholds. Here's what fails at 5, 15, and 30 clinicians — and the architecture gaps behind it.

Documentation burden in behavioral health: the hidden cost of manual notes
Manual therapy notes cost clinicians 6–8 unbillable hours every week. Here's what documentation burden actually costs your practice.