When patients first hear about Transcranial Magnetic Stimulation — or TMS — almost every one of them asks the same first question: "Does it actually work?"It's the right question to ask. TMS has been FDA-approved for major depression since 2008, with later approvals for OCD, anxious depression, smoking cessation, and adolescent depression. But "approved" and "actually works" are two different things. Patients want to know the realistic numbers — what percentage of people get better, how much better, and how long the improvement lasts.This article walks through what the evidence actually shows, in plain language, without overselling or underselling.
What "working" means in clinical terms
Before discussing numbers, it helps to know how clinicians measure improvement. Two terms come up repeatedly:Response means a patient experiences at least a 50% reduction in their depression symptoms, measured on a standardized rating scale. Response is meaningful but doesn't necessarily mean the patient feels fully well.Remission means depression symptoms have dropped below the clinical threshold — the patient is, by clinical measure, no longer in a depressive episode.A treatment can work in two ways: it can produce response (substantial improvement) or remission (full recovery). When researchers report effectiveness, they typically report both numbers, and they are different.This distinction matters because patients sometimes hear "TMS works in 60% of people" without knowing whether that means 60% feel somewhat better or 60% are fully well. The honest answer is somewhere in between, and it depends.
The numbers from peer-reviewed research
For major depression that has not responded to one or more antidepressant medications — what clinicians call treatment-resistant depression — published meta-analyses and large clinical trials consistently report:Response rate: approximately 50-60% of patients experience at least a 50% improvement in symptoms after a full TMS courseRemission rate: approximately 30-40% of patients achieve full remissionThese numbers come from controlled trials with rigorous methodology, including the open-label and randomized trials that supported FDA approval. A 2025 review in the American Journal of Psychiatry confirmed that roughly half of treatment-resistant patients achieve a clinically meaningful response to TMS.When TMS is combined with psychotherapy, the numbers improve further — some studies report response rates of 65-70% and remission rates of 50-55% with combined treatment.For OCD, deep TMS (a specific protocol FDA-approved in 2018) shows that more than one in three patients with treatment-resistant OCD achieve at least a 30% reduction in symptom severity. The OCD numbers are lower than the depression numbers, but they apply to a patient population that has typically failed multiple medications and CBT.
Why TMS works for some patients and not others
A 50-60% response rate also means 40-50% of patients don't respond meaningfully. Why? The honest answer is that we don't fully know in every individual case, but several factors are well-established.Patient factors:Severity and duration of depression at the start of treatmentNumber of previous medication failuresPresence of significant trauma history or co-occurring substance useIndividual neurobiology
— some brains respond more readily to neuromodulation than othersProtocol factors:Coil placement accuracy
— small differences in where the coil sits over the scalp matterStimulation intensity, calibrated to the individual's motor threshold (the lowest pulse strength that produces a small thumb movement when the coil is over the motor cortex)Frequency and protocol
— high-frequency, low-frequency, theta burst, and accelerated protocols all have different evidence bases for different conditionsClinic factors:Whether motor threshold is determined carefully and recalibrated through the courseWhether the protocol is adjusted if early sessions show no responseWhether psychotherapy and medication management are integrated alongsideThis is why the same machine and the same diagnosis can produce very different outcomes at different clinics. The protocol details and the clinical care around the protocol matter as much as the machine itself.
The realistic timeline of "working"
Patients sometimes expect to feel better after the first few sessions. That's not how TMS typically works. The realistic timeline:Sessions 1-5: Most patients feel nothing different mood-wise. The brain is being stimulated, but the cumulative effect needed for symptom change hasn't built up yet.Sessions 6-15: Many responders begin to notice changes around this window
— slightly better sleep, more interest in routine activities, less mental fog. The shifts are often subtle at first.Sessions 16-30: This is when the fuller response typically becomes apparent. Family members often notice the shift before the patient does.End of course (sessions 30-36): The full benefit of the protocol is evaluated. For responders, the improvement at this point is meaningful and sustained.A minority of patients
— sometimes called early responders — feel better within the first 10 sessions. Others need the full course before judging whether TMS has worked. This is one reason a partial course rarely gives a fair answer.
How long does the improvement last?
A common follow-up question: if TMS works, how durable is the improvement?Published data on six-month follow-up after a successful TMS course shows that more than half of patients who achieve response remain in response at six months. A meaningful proportion stay in remission for a year or longer. Others experience a gradual return of symptoms and benefit from periodic maintenance sessions — typically a session every one to three months, calibrated to the individual.Some patients do well with a single course and never need maintenance. Others use TMS as part of an ongoing treatment plan that includes therapy, medication, and lifestyle factors. Both patterns are normal.
What "not working" looks like — and what comes next
For the 40-50% of patients who don't respond meaningfully to a standard TMS course, the next clinical step depends on individual factors. Options include:
Switching to a different TMS protocol — for example, theta burst stimulation, or targeting a different brain region
Adding psychotherapy if it wasn't already part of the plan
Considering ketamine-assisted treatment or IV ketamine for treatment-resistant cases
Re-evaluating the medication regimen
For severe and refractory cases, considering ECT
The decision to call a TMS course "not working" should be made carefully. Sometimes a non-response is a partial response that needs more time. Sometimes it points to a missed underlying factor— sleep apnea, thyroid dysfunction, undisclosed substance use, an undiagnosed bipolar pattern.
A thorough re-evaluation before declaring failure is essential.
The honest answer
When patients ask "Does TMS actually work?", the honest answer is this: for about half of treatment-resistant patients, yes — and for many of those, the response is durable. For the other half, the response is partial or absent, and the next step depends on careful clinical reassessment.That is not the marketing answer, but it is the accurate one. TMS is a real treatment, with real evidence, that helps real patients. Patients deciding whether to start a course deserve realistic numbers and a clear-eyed clinician, not a brochure.The most important factor in whether TMS works for an individual patient is not the machine. It is the quality of the diagnostic evaluation that determines whether TMS is the right choice in the first place, the precision of the protocol, and the integration of TMS into a fuller treatment plan.