How many TMS sessions are needed?

A standard course is 36 — about 30 daily sessions over six weeks, then six tapering sessions over three more. That is what nearly every US clinic schedules and what the major insurers cover. It is a convention rather than a dose established by trial, but the largest observational analysis, covering 7,215 courses, found that people who finished with fewer than 30 sessions had worse outcomes than everyone else.

How long is a single TMS session?

It depends which stimulation pattern your clinic uses. Standard 10 Hz stimulation takes about 37.5 minutes. Intermittent theta burst, iTBS, takes about 3 minutes. A randomized trial of 414 patients found the two produced the same reduction in depression scores. Your first visit is longer than the rest, because it includes motor threshold mapping.

How many weeks does TMS take?

Six to nine on a standard schedule — six weeks of daily treatment, then a three-week taper. Accelerated protocols compress a larger number of sessions into about a week, but they are a different treatment plan with a thinner evidence base and different coverage, not a faster version of the standard one.

Can I have TMS more than once?

Yes, and roughly a third of people do — the device labeling reports that 36.2 percent required retreatment within twelve months. Coverage for a second course generally requires that the first one produced at least a 50 percent improvement which held for at least two months before you relapsed. If the first course did nothing, a second is usually not covered under those criteria.

Are maintenance or booster sessions covered?

Generally not. Both major coverage policies quoted on this page class maintenance TMS — anything outside the standard 30-plus-6 structure — as experimental or not medically necessary. Clinics do offer it. Expect to pay for it yourself, and ask for the per-session price before you commit to a course rather than after it ends.

Find a clinic

How many TMS sessions you need

Almost every clinic gives the same answer: 36 sessions, five days a week, over six to nine weeks. That answer is accurate. It is also a schedule that hardened into a rule, and the largest study of course length says the rule fits some people badly.

The short answer

A standard course is 36 sessions — around 30 daily treatments, then six tapering sessions spread over the final three weeks. Nearly every clinic and every major insurer works to that number. What is rarely said is where it came from. No randomized trial has ever compared a 36-session course against a 30-session or a 42-session one. The number settled into practice, and then into insurance policy, and the evidence about it arrived afterwards — from a registry of 7,215 real courses, which found that people who stopped short of 30 sessions did measurably worse, and that patients still improving at session 36 had not levelled off when their course ran out.

The standard course

What a standard course actually looks like

Most people arrive at TMS having been told a number without being told what it is made of. A standard acute course has two parts, and the second one surprises people: the taper is not extra treatment bolted on at the end, it is a wind-down, and on most schedules the last three weeks involve only a handful of visits.

Phase	Sessions	Schedule
Acute phase	30	One session a day, five days a week, for six weeks
Taper	6	Declining — several in the first week, one in the last — spread over three weeks
Total	36	About nine weeks from the first session to the last

The two-part structure is also where insurers quietly disagree with each other. Aetna covers "a maximum of 30 sessions (5 days a week for 6 weeks) plus 6 tapering sessions" — 36 in total, with the taper on top. Cigna says an initial course "typically includes 30-36 total treatments" and that "six tapered treatments over the final three-week period may be included in the total 30-36 visits" — with the taper counted inside the number. Read literally, the same phrase, a standard course, buys 30 daily treatments on one plan and as few as 24 on the other. Worth asking your clinic which count your authorization uses before the first session, not at week seven.

Where the number comes from

Half of every course ends on the same number

The best available evidence on course length is a 2023 analysis of a US treatment registry — 13,732 patients, of whom 7,215 had depression scores recorded both before and after treatment. Sorted by how many sessions people actually received, the distribution does not look like a biological dose. It looks like a reimbursement ceiling.

Share of 7,215 TMS courses, by number of sessions received · US treatment registry · track ends at 50%

1 to 19 sessions
658 patients · 9.1%
20 to 29 sessions
616 patients · 8.5%
30 to 35 sessions
1,375 patients · 19.1%
Exactly 36 sessions
3,591 patients · 49.8%
37 to 41 sessions
626 patients · 8.7%
More than 41 sessions
349 patients · 4.8%

Three things are true about that number at once, and clinic pages tend to state only the first. It is not in the device labeling — the FDA-cleared prescribing information for NeuroStar, the first system cleared for depression, names no session count at all. It is not the trial number either: the authors of the largest analysis of TMS dosing open their paper by noting that the number of sessions in an acute course is now greater than in the earlier randomized controlled trials. And nobody has randomized anyone to a course length. The studies that exist compare pulses per session, sessions per day, and one stimulation pattern against another. None of them assign patients to 30 sessions or 36 or 42 and compare what happens.

Running short, running long

What the registry found about stopping early and going long

Three findings came out of the same 7,215 courses, and they do not all point the same way.

Under 30 sessions did worse

Both short-course groups had "inferior endpoint outcomes than all other groups." Nearly one patient in five in this registry — 1,274 of 7,215 — finished with fewer than 30 sessions.

The 36-session group improved most

Symptom reduction was greatest in the group that ended treatment at exactly 36. That is also the group whose course was least likely to have been cut short for a clinical reason, which is worth holding in mind when reading it.

Going past 36 kept working

Patients who ran long showed "less antidepressant response early in the course" and "a slower but steady rate of improvement over time." Extending beyond 36 was "associated with further improvement without evidence of a plateau."

That last finding is the one to hold on to. In the largest dataset anyone has, the people still improving at session 36 were on average the people who had improved least by session 10, and their curve had not flattened when the course ended. It is exactly the group whose insurer is most likely to ask why they need more.

Who sets the ceiling

Your insurer decides how many sessions you get

Not your clinician alone. Coverage policy sets the ceiling, and the two policies quoted below are public documents you can read yourself. They do not agree with each other, and neither of them agrees with the largest study of course length.

Covers "a maximum of 30 sessions (5 days a week for 6 weeks) plus 6 tapering sessions (6 sessions over three weeks)." Treatments beyond 36 "may be reviewed for medical necessity," on the stated grounds that evidence is insufficient that additional sessions help delayed responders reach remission. Maintenance TMS — anything outside that structure — is classed as experimental and is not covered.
Aetna Clinical Policy Bulletin 0469

An initial course "typically includes 30-36 total treatments, with one treatment per day, over a 4-6-week period," with "six tapered treatments over the final three-week period" counted inside that total rather than added to it. TMS "used as a maintenance therapy, is considered not medically necessary."
Cigna Medical Coverage Policy 0383

Set the Aetna sentence beside the registry finding and the conflict is exact. The policy says there is insufficient evidence that going past 36 helps the people who respond slowly. The largest study of course length found that the people who go past 36 are specifically the slow responders, that they keep improving, and that their improvement had not plateaued when treatment stopped.

Both positions are defensible, which is the awkward part

The policy is asking for randomized evidence, and randomized evidence about course length does not exist. The registry is observational, and patients whose clinicians chose to extend their course are not a random sample of anyone. But the practical result is that the group with the most to gain from a longer course is the group most likely to have one declined, and the reason is an absence of trials rather than a finding from one.

Two things worth asking before you need them

Which count does my authorization use — 30 plus a taper, or 30 to 36 including it? And if my course needs extending, what does the clinic submit? Documented response to date is the material a reviewer reads, and it is far easier to collect from session one than to reconstruct at session 30.

Time in the chair

A session is not an appointment

Thirty-six sessions tells you nothing about how much of your life this takes, because two very different treatments are both called a session — and clinics quoting a session count rarely say which one they mean.

37.5 min

is a standard 10 Hz session — about 22.5 hours of stimulation across a full course

3 min

is an intermittent theta burst session, and a trial of 414 people found the two equivalent

36 trips

on 36 weekday mornings is the cost most people underestimate, whichever pattern is used

The trial that settled this randomized 414 patients to one of two stimulation patterns, five days a week for four to six weeks. Depression scores fell by the same amount in both arms — 23.5 to 13.4 against 23.6 to 13.4 on the 17-item Hamilton scale, an adjusted difference of 0.103 points, which is how theta burst was established as non-inferior to the older, longer protocol. Twenty hours of difference in chair time, same measured result. It is a reasonable thing to ask a clinic about.

Two things the numbers do not capture. Your first visit is longer than the rest, because it includes motor threshold mapping — the clinician finds the stimulation level that makes your thumb twitch and sets your dose from it. And missed sessions are the common worry with no honest answer available, because no trial has measured what a gap does to the outcome. What is known is only the registry finding above, that courses ending under 30 sessions did worse. That is an argument for completing the course, not for panicking about one Tuesday.

Accelerated protocols

Compressed courses change the calendar, not the count

Accelerated protocols are the most-searched thing in this area and the most misunderstood. They do not reduce the number of sessions. They increase it, and compress it into days.

The Stanford accelerated protocol (SAINT)	As published
Sessions per day	10
Days	5 consecutive
Total sessions in the course	50
Gap between sessions	50 minutes

Protocol as published in the 2020 Stanford trial. Cleared by the FDA in 2022.

That is 50 sessions inside a working week, against 36 spread over nine. Each session carries 1,800 pulses, for 90,000 across the week — roughly five times the pulse count of a standard six-week theta burst course. So the calendar shrinks and the dose goes up. It answers a different question from the one this page is about.

The published remission figure is striking and the study behind it is small: 21 patients in the per-protocol analysis, 19 of whom met remission criteria, in an open-label design with no sham control. Later controlled work has produced lower numbers. If a clinic quotes you 90 percent, that figure comes from those 21 people, and it is worth asking what the controlled results were.

After the course

When 36 sessions is not enough

Roughly half the people who complete a course do not get a meaningful response. That figure, and where it comes from, is on our page on whether TMS works. What belongs here is what happens next — because the rules are specific, and most people meet them at the worst possible moment.

Covered, if the first course worked

A second course after a relapse — both policies require at least a 50% symptom reduction during the previous course
That improvement must have been maintained for at least two months before the relapse
Aetna adds that a repeat course within 60 days of the last one is not medically necessary
Extending a course past 36 sessions, where the clinic can document response to date
Retreatment is common — the labeling reports 36.2% of patients needed a further course within twelve months

Not covered, or not settled

Maintenance or booster sessions — both policies class ongoing maintenance TMS as experimental or not medically necessary
A second course after a first one that produced nothing — the retreatment criteria describe relapse, not rescue
Any session count beyond 36 as an entitlement rather than a review
What a gap between sessions costs you — no study has measured it

Read the left column carefully, because it is a relapse clause rather than a rescue clause. If your first course produced nothing, the retreatment criteria in these policies do not describe you, and a second course becomes a conversation about exceptions instead of a covered benefit. Relapse itself is common enough for the rule to bite: the labeling reports that 29.5 percent of people who reached remission relapsed within the year. If you are mid-course and wondering whether to stop rather than looking back at a course that finished, that is a different question, and our page on the TMS dip is the one to read.

If this is getting worse, not just slower

A course that is not working is not the same thing as an emergency, but the two can arrive together. New or worsening thoughts of harming yourself during a course are something your clinic needs to hear about today, not at your next scheduled visit.

If you are in crisis right now, call or text 988 in the US to reach the Suicide and Crisis Lifeline, or text HOME to 741741 for the Crisis Text Line. Both are free and open 24 hours. Call 911 for a seizure or any medical emergency during or after a session.

FAQ

The questions people ask about this

Six questions, answered against the same sources as the rest of the page.

How many TMS sessions are needed?
A standard course is 36 — about 30 daily sessions over six weeks, then six tapering sessions over three more. That is what nearly every US clinic schedules and what the major insurers cover. It is a convention rather than a dose established by trial, but the largest observational analysis, covering 7,215 courses, found that people who finished with fewer than 30 sessions had worse outcomes than everyone else.
How long is a single TMS session?
It depends which stimulation pattern your clinic uses. Standard 10 Hz stimulation takes about 37.5 minutes. Intermittent theta burst, iTBS, takes about 3 minutes. A randomized trial of 414 patients found the two produced the same reduction in depression scores. Your first visit is longer than the rest, because it includes motor threshold mapping.
How many weeks does TMS take?
Six to nine on a standard schedule — six weeks of daily treatment, then a three-week taper. Accelerated protocols compress a larger number of sessions into about a week, but they are a different treatment plan with a thinner evidence base and different coverage, not a faster version of the standard one.
What if I miss a session?
No study has measured the effect of a missed session or a gap, so nobody can honestly give you a number. What the evidence does show is that courses ending under 30 sessions did worse than longer ones, which is a reason to complete the course rather than a reason to panic about one absence. Tell your clinic and reschedule.
Can I have TMS more than once?
Yes, and roughly a third of people do — the device labeling reports that 36.2 percent required retreatment within twelve months. Coverage for a second course generally requires that the first one produced at least a 50 percent improvement which held for at least two months before you relapsed. If the first course did nothing, a second is usually not covered under those criteria.
Are maintenance or booster sessions covered?
Generally not. Both major coverage policies quoted on this page class maintenance TMS — anything outside the standard 30-plus-6 structure — as experimental or not medically necessary. Clinics do offer it. Expect to pay for it yourself, and ask for the per-session price before you commit to a course rather than after it ends.

Sources

Every number on this page

Grouped by what they support. Every entry links out to the paper, the device labeling or the coverage policy itself, not to a page summarising it.

The pillar, and the two closest siblings

Pillar

What is TMS therapy?

Spoke

What to expect during TMS treatment