For providers

90832, 90834, 90837: the time bands and where the minutes go

Summary

The three individual-psychotherapy codes are time bands: 90832 covers roughly 16 to 37 minutes, 90834 covers 38 to 52, and 90837 covers 53 minutes and up. You choose the code whose range holds the actual face-to-face therapy time — the service itself, not the whole appointment. Under 16 minutes is not separately reportable. The band has to match the time your note documents, which is why start-and-stop times matter and why the top band draws the most payer scrutiny.

By Gale Editorial · Updated 2026-07-26. Every figure cited to a dated source. How we write.

The three codes and their time bands

The three individual-psychotherapy codes are defined by time, and each one is a band. 90832 is the code for roughly 16 to 37 minutes of psychotherapy, 90834 for 38 to 52 minutes, and 90837 for 53 minutes or more 1. You do not pick among them by intensity, difficulty, or content — you pick the one whose time range contains the actual face-to-face psychotherapy time delivered.

The three describe the same service at three lengths. CPT frames each with a typical time — think 30, 45, and 60 minutes — surrounded by a range, and it is the range that governs which code goes on the claim 1. A session that runs a genuine 45 minutes is 90834 whether it was routine or intricate; a session that runs a genuine 60 minutes is 90837 for the same reason. The clinical weight of the hour does not move the code; the clock does.

There is a floor beneath all three. Below 16 minutes of psychotherapy, no timed psychotherapy code is separately reportable 1. A brief check-in that does not reach the floor is not a 90832 rounded down — it is a service that the timed psychotherapy codes do not describe. Getting the floor right matters as much as getting the bands right, because billing a timed code for time that was never spent is the error a review is built to catch.

Where the minutes actually come from

The minutes that decide the band are the psychotherapy minutes, not the length of the appointment. Time spent on the therapy service itself is what counts toward 90832, 90834, or 90837 1 — not the scheduling, not the walk from the waiting room, not the note written after the patient leaves. A fifty-minute slot on the calendar is not automatically a 90837, and treating the appointment length as the code is one of the most common ways a claim drifts out of line with the record.

The boundaries between the bands are real and narrow. A session of 52 minutes of psychotherapy is 90834; a session of 53 minutes is 90837. That one-minute difference is why the note, not the schedule, has to establish which side of the line the session landed on. The published ranges already encode the convention that you reach a code once you pass the midpoint toward it, so there is no separate rounding rule to apply — the range is the rule.

This is also why "about an hour" is not a code. The band is chosen from the documented psychotherapy time, and a habit of billing the same code for every session regardless of its actual length is exactly the pattern that invites a request for the underlying times. The safest practice is the literal one: count the therapy minutes, and bill the band that contains them.

Start and stop times: what the chart has to show

Because the code follows the clock, the chart has to show the clock. The defensible way to support a timed psychotherapy code is a note that records the session's actual duration — start and stop times, or total psychotherapy minutes — so the band billed matches the time documented 1. When a payer questions a timed code, the recorded time is the first thing it looks for, and its absence is what turns a routine question into a denial.

A common convention is to record start-and-stop times as a standard field on every psychotherapy note, not only on the long ones. It costs nothing at the time and it converts a code-selection judgment into a documented fact. A note that says "53 minutes, 3:02 to 3:55" answers the only question the band raises; a note that says "session held" leaves the code unsupported no matter how accurate it was.

The discipline pays off unevenly but reliably. Most sessions never get reviewed, and for those the times are simply good record-keeping. The occasional session that does get reviewed is the one where the documented time is the difference between a code that stands and a code that gets recouped. Recording the time on all of them is cheaper than sorting later which ones will be questioned.

90837 and the audit spotlight

90837, the top band, draws more payer attention than the other two, and it is worth understanding why before it becomes a string of denials. It is the highest-paying of the three timed individual codes and the one most associated with utilization review, so a practice whose claims are heavily weighted toward it can expect questions about time and medical necessity in a way a mixed pattern does not attract.

This is 90837 under the microscope, and the framing that helps is that the scrutiny is about the pattern, not the individual code. There is nothing improper about billing 90837 for a genuine 53-plus-minute session — that is what the code is for. What draws review is a distribution that looks like every session is the longest possible one, which a payer reads as a prompt to check whether the times were real.

The defense is the same discipline that supports any timed code, applied consistently: a note that documents the actual duration and the clinical reason the session ran as long as it did. A practice that records start-and-stop times and writes notes that show why the work took the time it took has already built its answer before the question arrives. The clinicians who struggle with 90837 review are usually the ones whose records cannot reconstruct the time after the fact, not the ones who saw long sessions. A quick self-check helps: if a payer asked tomorrow for the times behind a month of 90837 claims, could the notes produce them? If the answer is yes, the pattern takes care of itself.

When psychotherapy rides on top of a med visit

When a prescriber or other qualified clinician provides psychotherapy during a medical evaluation-and-management visit, the two services are coded separately rather than merged. The E/M visit is reported on its own, and the psychotherapy is added with a psychotherapy add-on code selected by its own time 1. The add-on codes correspond to the same time increments as the standalone therapy codes, and the therapy minutes are carved out and counted only once.

The interaction with the E/M level is the part that trips people. When a psychotherapy add-on is billed alongside an E/M service, the E/M level is selected on medical decision-making rather than on total time, because the psychotherapy minutes cannot also count toward the E/M 1. In other words, the two clocks do not overlap: therapy time supports the add-on, and the medical decision-making supports the E/M.

That rule sits inside the broader 2021 office-visit framework, under which office and outpatient E/M levels are chosen by medical decision-making or total time, and history and exam no longer set the level 2. The medical decision-making table and the documentation each level expects are spelled out in Medicare's evaluation-and-management guidance 3. For a solo who both prescribes and does therapy, the practical takeaway is to document the two services as two services — the medical decision-making that drives the E/M, and the separately timed psychotherapy that drives the add-on. A note that blurs them into one narrative gives a reviewer no way to see two distinct services, and the add-on is the piece that tends to fall out under scrutiny when the psychotherapy time and content are not set apart from the medical portion.

Who sets these codes and why they change

These codes are not payer inventions, which is why they behave consistently across payers. They are CPT, maintained by the American Medical Association's CPT Editorial Panel, updated annually, and named as the standard procedure code set for professional services 4. The definitions of 90832, 90834, and 90837 are the same whether the claim goes to Medicare or a commercial plan; what varies is the rate a given payer attaches to each, not the meaning of the code.

Because CPT updates on an annual cycle, the codes and their descriptors are worth a look each year rather than assumed to be frozen. Most years the psychotherapy family is stable, but the surrounding rules — coverage, who may bill, telehealth treatment — shift more often than the code definitions themselves.

Medicare coverage is one of those moving pieces. Medicare covers the psychiatric diagnostic evaluation and psychotherapy services, and since 2024 marriage and family therapists and mental health counselors can enroll in Medicare and bill for these services directly — a change that widened who may report the psychotherapy codes to the program 5. A clinician who was previously locked out of Medicare billing by profession is worth re-checking against the current rules, because the answer changed recently and may have changed in their favor.

The neighbors: intake, family, and group

The individual time bands stop where other services begin, and knowing the boundary keeps the wrong code off the claim. The psychiatric diagnostic evaluation — the intake — is 90791, and it is not selected by the same minute bands as the timed therapy codes 1. The long intake is its own service with its own rules, not a stretched 90837, and billing the intake as an individual therapy band because it ran long is a category error rather than a close call.

Family and group work also live outside the individual bands. Family psychotherapy is 90846 when the patient is not present and 90847 when the patient is present, and couples work is generally reported under the family codes rather than an individual time band 1. Group psychotherapy is 90853. Each of these is a distinct service with its own documentation and its own place on the claim.

The unifying point is that the time bands answer exactly one question — how long was the individual psychotherapy session — and they answer nothing else. When the service is an evaluation, a family session, or a group, the individual bands are the wrong tool, and reaching for one to describe a different service is a coding mistake, not an efficiency. Match the code to the service first, then let the clock choose the band only when the service is individual psychotherapy.

Common questions

90832 covers roughly 16 to 37 minutes of individual psychotherapy, 90834 covers 38 to 52 minutes, and 90837 covers 53 minutes or more. You bill the code whose range contains the actual face-to-face therapy time. Below 16 minutes, no timed psychotherapy code is separately reportable, so a very brief contact is not a rounded-down 90832.

No. Only the psychotherapy service itself counts toward the band — not scheduling, not waiting-room time, not documentation completed after the patient leaves. A fifty-minute appointment slot is not automatically a 90837. The boundaries are narrow: 52 minutes of therapy is 90834 and 53 is 90837, so the note, not the calendar, has to establish which side of the line the session landed on.

It is the highest-paying of the three timed individual codes and the one most associated with utilization review. What draws attention is usually the pattern — claims weighted heavily toward the longest band — rather than any single use. There is nothing improper about billing 90837 for a genuine 53-plus-minute session; the defense is documenting the actual time and the clinical reason the session ran that long.

Code the two services separately. The E/M visit is reported on its own, and the psychotherapy is added with a psychotherapy add-on code chosen by its own time. When the add-on is billed with an E/M service, the E/M level is selected on medical decision-making rather than total time, because the psychotherapy minutes cannot also count toward the E/M. The two clocks do not overlap.

Yes, as of 2024. Marriage and family therapists and mental health counselors can now enroll in Medicare and bill directly for psychiatric diagnostic evaluation and psychotherapy services, including the timed individual codes. A clinician previously locked out of Medicare billing by profession is worth re-checking against the current rules, since this changed recently.

Run your practice on Gale

The software is free. Gale earns one flat 3.5% all-in per paid transaction — only on transactions that actually pay. No subscription, no setup fee, no network cut.

Start or manage a practice →

References

  1. 1.APA Services, Inc. (2025). Psychotherapy Codes for Psychologists. APA Services, Inc.. link ✓That 90832, 90834, and 90837 are time-banded individual-psychotherapy codes selected by the face-to-face psychotherapy time (roughly 16-37, 38-52, and 53+ minutes) with a 16-minute floor; that psychotherapy add-on codes report therapy layered onto an E/M visit by their own time; and that 90791 intake, 90846/90847 family, and 90853 group are distinct services outside the individual bands.
  2. 2.American Medical Association (2023). CPT evaluation and management (E/M) revisions. American Medical Association (AMA). link ✓That the 2021 framework selects office and outpatient E/M levels by medical decision-making or total time, and eliminated history and exam as level-setting elements.
  3. 3.Centers for Medicare & Medicaid Services (2023). Evaluation and Management Services Guide. CMS Medicare Learning Network (MLN006764). link ✓That the medical decision-making table and the documentation each E/M level expects are set out in CMS's evaluation-and-management guidance.
  4. 4.American Medical Association (2026). CPT® (Current Procedural Terminology). American Medical Association (AMA). link ✓That CPT is maintained by the AMA CPT Editorial Panel, updated annually, and is the standard (HIPAA-named) procedure code set for professional services.
  5. 5.Centers for Medicare & Medicaid Services (2025). Medicare and Mental Health Coverage. CMS Medicare Learning Network (MLN1986542). link ✓That Medicare covers psychiatric diagnostic evaluation and psychotherapy and, since 2024, allows marriage and family therapists and mental health counselors to enroll and bill for these services. As of July 2026.

https://www.gale.care/for-providers/bhc-psychotherapy-time-bands · 5 sources. Competitor details are cited to dated public sources and maintained as they change; figures are estimates, not commitments. Synthetic demonstration.

← Back to for providers

Findability, by specialty

How practices like yours get found in local search and AI answers — the honest playbook, per specialty.

SEO for private practices · SEO for AI search / answer engines (all verticals)