In clinical practice, roughly 60 to 75 percent of patients with treatment-resistant depression respond to a course of TMS and roughly 30 to 50 percent reach remission, against a 10 to 15 percent remission rate for a third or fourth antidepressant. Controlled trials show lower but significant rates versus sham. Results hold at one year for most responders.
The context matters. Everyone who reaches TMS has already tried medication without enough benefit, which means they belong to the group of patients for whom the next medication is least likely to work. Measured against that baseline, response rates of roughly 60 to 75 percent and remission rates of roughly 30 to 50 percent in clinical practice are a meaningful improvement over the alternatives. They are also not a guarantee, and we tell every patient that before they start.
Who it is for
This page is for anyone weighing TMS against another medication, Spravato or waiting, and for anyone who has seen a clinic advertise a 90 percent success rate and wants to know whether to believe it. The figures apply to adults with major depressive disorder who have not responded to at least one, and usually two or more, antidepressant trials, which describes nearly every patient treated with TMS. We also cover the OCD results separately, since that protocol and its outcomes differ.
How it works
Depression outcomes are measured with standardized scales: the PHQ-9 that patients complete themselves, and clinician-rated scales such as the Hamilton Depression Rating Scale (HAM-D) and the Montgomery-Asberg Depression Rating Scale (MADRS). A response means the score fell by at least half from baseline. Remission means the score fell below a cutoff that corresponds to being essentially well, for example a HAM-D of 7 or less or a PHQ-9 below 5. Controlled trials compare active TMS to a sham coil that mimics the sound and feel without stimulating the brain, so the active figure must be read against the sham figure. Real-world registries track outcomes in ordinary clinic patients without a sham group and with patients continuing their medications, which produces higher absolute numbers for reasons discussed below.
In STAR*D, about 37 percent of patients remitted on the first antidepressant, about 31 percent on the second step, and only around 13 to 14 percent on the third and fourth steps, with higher relapse rates at each step.
Source: Rush et al., American Journal of Psychiatry, 2006
In the pivotal multicenter randomized sham-controlled Deep TMS trial of 212 patients, remission at week five was 32.6 percent with active treatment versus 14.6 percent with sham, and response was 38.4 percent versus 21.4 percent.
Source: Levkovitz et al., World Psychiatry, 2015
In a large naturalistic outcome study of TMS in routine clinical practice, 58 percent of patients responded and 37 percent remitted on clinician-rated scales, with higher rates on self-report measures.
Source: Carpenter et al., Depression and Anxiety, 2012
In the randomized sham-controlled Deep TMS trial for OCD, 38 percent of patients receiving active treatment met the response criterion at six weeks compared with 11 percent receiving sham, rising to 45 percent versus 18 percent one month later.
Source: Carmi et al., American Journal of Psychiatry, 2019
What to expect
If you are treated with Deep TMS at Stedfast Psychiatry, the odds based on the published evidence are roughly as follows. Two in three patients will have a meaningful response by the end of the course. About four in ten will reach remission. Of those who respond, roughly two thirds will still be responders a year later, and about one third will need a repeat course at some point during that year. Patients who have failed fewer medications, who are earlier in the current episode, and who continue an antidepressant alongside TMS tend to do better. We track your PHQ-9 weekly, and an improvement of 20 percent or more by session 10 to 15 is a favorable sign, though late responders exist and we do not stop early on that basis alone.
Benefits
- Response and remission rates two to three times higher than a further antidepressant trial in the same patients
- Efficacy demonstrated against sham in randomized multicenter trials, not only open-label data
- Benefit for anxiety symptoms alongside depression in pooled trial analyses
- A separate randomized trial supporting the OCD protocol
- Durability at 12 months for the majority of responders
- A repeat course works for most patients who responded the first time
Risks and side effects
The main risk relevant to this page is the risk of non-response: roughly one quarter to one third of patients do not get meaningful benefit from a course of TMS, and a smaller group get partial benefit that is not enough to change their lives. We cannot yet predict with certainty who those patients will be. Side effects are covered on our safety page; in brief, scalp discomfort and headache are common and transient, seizure is rare, and there are no cognitive side effects, which is a point of contrast with ECT.
Cost and insurance
Insurers cover TMS because the evidence summarized here met their standards for medical necessity. Medicare and most commercial plans in Florida cover a course for major depressive disorder after documented medication failure, and most plans will cover a repeat course for patients who responded previously and have relapsed. Coverage for the OCD protocol is less uniform and our coordinator checks it plan by plan. The expected outcomes also inform the economics: a course that produces remission in a patient who has cycled through four medications usually costs less over the following year than continued non-response.
The baseline: what STAR*D showed about medication after medication
The Sequenced Treatment Alternatives to Relieve Depression study enrolled more than 4,000 outpatients and treated them through up to four successive steps. On the first medication, citalopram, about 37 percent remitted. On the second step, about 31 percent of those still depressed remitted. On the third and fourth steps, remission fell to around 13 to 14 percent each. Cumulatively about two thirds of patients eventually remitted, but the later the step, the higher the relapse rate in the following year. Patients who arrive for TMS have typically completed step two, three or four. The honest comparison for TMS is therefore not against a first antidepressant but against the 13 to 14 percent remission that another medication would offer at that point.
The controlled evidence for Deep TMS in depression
The pivotal trial published in World Psychiatry in 2015 randomized 212 patients at 20 centers in the United States, Canada, Israel and Germany to active Deep TMS or sham. All had failed one to four antidepressants in the current episode and were off medication during the study. After four weeks of daily sessions and a further 12 weeks of twice-weekly sessions, the primary analysis at week five showed a remission rate of 32.6 percent with active treatment against 14.6 percent with sham, and a response rate of 38.4 percent against 21.4 percent. The advantage persisted at week 16. These results, along with the earlier figure-8 coil trials, are the basis for the FDA clearance and for insurer coverage. They are modest-looking numbers because they are measured against a sham that itself produced some improvement and because patients were off medication, which is not how TMS is delivered in practice.
Real-world outcomes in clinics
Registry and naturalistic studies measure what happens to ordinary patients treated in ordinary clinics, most of them continuing their antidepressants. The largest early study, published in Depression and Anxiety in 2012, followed 307 patients across 42 practices and found clinician-rated response in 58 percent and remission in 37 percent, with higher figures on self-report. Later and larger registries, including post-marketing data on Deep TMS, report response rates in the range of 60 to 75 percent and remission in the range of 30 to 50 percent depending on the scale and completion criteria used. We quote that range to patients. It reflects the treatment as delivered, with medication continued and courses completed, and it has been consistent across multiple independent data sets.
OCD: a separate protocol and a separate trial
Deep TMS for OCD uses a different coil that targets the medial prefrontal and anterior cingulate cortex, and each session begins with a brief, individualized provocation of obsessional anxiety so that the relevant circuit is active during stimulation. The randomized sham-controlled trial, published in the American Journal of Psychiatry in 2019, enrolled 100 patients who had failed SSRIs and, in most cases, exposure therapy. At six weeks, 38 percent of the active group had a response, defined as a 30 percent or greater drop in Y-BOCS, versus 11 percent with sham, and at one-month follow-up the figures were 45 percent versus 18 percent. The average Y-BOCS reduction was about six points with active treatment. These are meaningful gains in a highly resistant population, and they are partial gains: TMS for OCD reduces symptoms rather than eliminating them and works best alongside continued exposure and response prevention therapy.
Durability: how long the benefit holds
The most cited one-year study, published in the Journal of Clinical Psychiatry in 2014, followed 257 patients after an acute course of TMS. Among those who had responded, 62.5 percent maintained response throughout the 12 months, and 36.2 percent received a repeat course at some point, usually with good results. A 2019 meta-analysis of durability studies found sustained response in about 66 percent of responders at three months, 53 percent at six months and 46 percent at 12 months without maintenance treatment. The Deep TMS trial itself showed the benefit holding through 16 weeks of continuation sessions. The practical reading is that most responders stay well for a year, a substantial minority relapse and need a repeat course, and continuing medication and psychotherapy after TMS tilts the odds toward durability.
Predictors of response
| Factor | Effect on response |
|---|---|
| Fewer failed antidepressants in the current episode | Favorable; the strongest predictor |
| Shorter current episode | Favorable |
| Younger age | Modestly favorable, but older adults still respond well |
| Continuing an antidepressant during TMS | Favorable |
| Early improvement by session 10 to 15 | Favorable, though late responders exist |
| High-dose benzodiazepine use | Unfavorable |
| Psychotic features or active substance use | Unfavorable; usually addressed first |
| Prior response to TMS | Strongly favorable for a repeat course |
The honest limits
Roughly a quarter to a third of patients do not respond meaningfully to TMS, and we cannot reliably predict who they will be in advance. Sham response in trials is real, which means some of the improvement any patient feels is not specific to the stimulation. Most trial evidence covers four to six weeks of acute treatment and up to a year of follow-up; longer-term data are thinner. Evidence for conditions outside the FDA-cleared indications, including bipolar depression, PTSD and chronic pain, is promising but preliminary. And TMS requires a daily commitment for six to nine weeks that not every patient can make. None of this changes the central finding: for patients who have not responded to medication, TMS is among the best-supported next steps available, and its results are good enough to state plainly.
Deep TMS at Stedfast Psychiatry, Palm Beach Gardens
Stedfast Psychiatry delivers Deep TMS at our Palm Beach Gardens office, 11641 Kew Gardens Ave, Suite 209, serving patients from across Palm Beach, Martin and St. Lucie counties. We track every patient's PHQ-9 weekly and will show you your own trend line, not just the published averages. Call (561) 437-2199 to schedule a TMS consultation and get a candid assessment of your likely odds based on your history, or ask your current Stedfast provider for a referral.
Frequently asked questions
What is the success rate of TMS for depression?
In routine clinical practice, published registries report that roughly 60 to 75 percent of patients respond, meaning their depression score falls by at least half, and roughly 30 to 50 percent reach remission. In randomized trials against a sham coil, the active rates are lower, about 38 percent response and 33 percent remission in the pivotal Deep TMS trial, but still more than double the sham rates. Both sets of numbers are honest; they measure different things, and both should be read against a 10 to 15 percent remission rate for a third or fourth antidepressant.
Why are the real-world numbers higher than the trial numbers?
Several reasons. Trials require patients to stop antidepressants, while clinic patients continue them, and the combination works better. Trials use strict entry criteria and short fixed courses; clinics treat a broader population and can extend a course for slow responders. Trial outcomes are measured at a fixed endpoint by blinded raters; registries often use self-report scales, which show larger changes. And there is no sham group in a registry to subtract. The trial numbers prove the treatment works; the registry numbers estimate what to expect in a clinic.
What does STAR*D have to do with TMS?
STAR*D is the largest study of sequential antidepressant treatment ever conducted, and it established the baseline against which TMS is judged. About a third of patients remitted on the first medication. Each subsequent step helped fewer people: by the third and fourth medications, remission was around 13 to 14 percent, and those who did remit relapsed more often. Patients referred for TMS are typically at that third or fourth step. A treatment that remits 30 to 50 percent of them is a large improvement over the next pill.
How well does TMS work for OCD?
The Deep TMS OCD protocol, which uses a different coil target and includes brief symptom provocation before each session, was tested in a randomized sham-controlled trial published in the American Journal of Psychiatry in 2019. At six weeks, 38 percent of active-treatment patients had at least a 30 percent reduction in Y-BOCS score compared with 11 percent with sham, and the gap widened at one-month follow-up. Those were patients who had already failed SSRIs and exposure therapy. OCD responses are typically partial rather than complete, and TMS works best alongside continued therapy.
How long do the results last?
In the largest one-year follow-up study, about two thirds of patients who responded to TMS were still responders at 12 months, and about one third received a repeat course at some point during the year, most of whom responded again. A meta-analysis of durability studies found sustained response in roughly two thirds of responders at three months and about half at one year. Continuing an antidepressant and psychotherapy after TMS improves durability. Our maintenance page covers the options for patients whose symptoms begin to return.
What makes someone more likely to respond?
The strongest predictor is how many antidepressants have failed: patients who have failed one or two respond more often than those who have failed five or six. Shorter duration of the current episode, younger age, continuing an antidepressant during TMS, completing the full course without long gaps and early improvement by the midpoint all favor response. High-dose benzodiazepines appear to reduce response. Psychotic depression and active substance use predict poorer outcomes. None of these is absolute; patients with every unfavorable factor still respond sometimes.
Can I trust clinics that advertise 80 or 90 percent success?
Be skeptical. Those figures usually come from counting any improvement on a self-report scale, from small samples, or from excluding patients who dropped out. The highest-quality real-world data put response around 60 to 75 percent and remission around 30 to 50 percent, and the randomized trials are lower. A clinic that quotes the range we quote here, and tells you up front that a quarter or more of patients do not respond, is giving you numbers you can plan around.
What happens if TMS does not work for me?
You still have options, and we will have been tracking your scores closely enough to know by the end of the course. Spravato (esketamine) nasal spray is FDA-approved for treatment-resistant depression and is available at three of our Florida offices; it works through a different mechanism and some TMS non-responders do well on it. Augmentation strategies, a switch in antidepressant class, and ECT for severe depression remain available. Partial responders sometimes benefit from an extended course. Non-response to TMS is disappointing, but it is not the end of the road.
Where this treatment is offered
Conditions this treats
If you are in crisis, call or text 988 or go to the nearest emergency room. Educational content; individual results vary. Reviewed by Dr. David Husted, board-certified psychiatrist.