How to choose: finding the right fit

When medication has not been enough, there is rarely one “best” treatment; there is a best fit for your situation. This page explains what “treatment-resistant” means, how clinicians weigh TMS, ketamine, esketamine, ECT and VNS, what the guidelines say, what is not yet approved, and the questions worth bringing to a consultation. Current to October 26, 2026.

Finding the right fit

If you or someone you love has depression that has not improved with the usual treatments, you have probably run into a lot of competing claims about what to do next. This page is written for patients and families. It explains what “treatment-resistant” actually means, how a clinician weighs the choice among more medication, psychotherapy, TMS, ketamine and esketamine, ECT and vagus nerve stimulation, what the research says in plain numbers, and how the decision gets made with you rather than for you.

Two things are true at the same time. Some of these treatments work better than others on average. And “best on average” is not the same as “right for you.” Your health, your history, how urgent things are, what you can fit into your life and what you are willing to accept all change the answer. My job is not to tell you what to do. It is to lay out the real options and the real numbers, including the people who do not get better, and then to decide together.

Where the evidence stands on October 26, 2026:
  • Most definitions of treatment-resistant depression require two adequate courses of antidepressant medication that did not work, but there is no single agreed definition, and a significant share of people labelled “resistant” never had an adequate trial.1,2
  • In a large US government-funded study of step-by-step medication treatment (STAR*D, 4,041 people enrolled), the investigators reported a cumulative remission rate of 67% across four steps, a projected figure that assumed people who dropped out would have done as well as those who stayed; a 2023 reanalysis using the study’s own protocol rules found 35%.3,4
  • ECT has the largest treatment effect of any brain-stimulation treatment in head-to-head and network comparisons (with wide confidence intervals), and more memory side effects than TMS or ketamine.5,6,7,8,9
  • In a US trial of people without psychosis, ketamine infusions were not worse than ECT (response 55% versus 41%); in a Swedish inpatient trial, ECT did better (remission 63% versus 46%). Which result is closer to your situation may depend on how ill you are and whether you are treated in hospital; that pattern comes from subgroup analyses, not from a trial designed to test it.9,10,11,12
  • Decision aids improved knowledge and reduced feeling uninformed across 209 trials, but the one decision aid built for treatment-resistant depression has only been piloted in five people, and we found no randomized trial of a decision aid for choosing among TMS, ketamine, ECT or VNS.13,14
  • Psilocybin and MDMA are not FDA-approved for depression.15,16,17

This page is about how to choose. The treatments themselves each have their own page: TMS, ketamine and esketamine, ECT, vagus nerve stimulation and psychedelics.

What “treatment-resistant” means

“Treatment-resistant depression” sounds like a diagnosis. It is really a label, and there is no single agreed definition behind it. A 2020 review done for Medicare and the Agency for Healthcare Research and Quality found no consensus. The most common definition required at least two prior treatment failures, with confirmation that each medicine was taken at an adequate dose for an adequate length of time. Only 17% of the treatment studies the reviewers examined had actually enrolled people who met that definition.1

The FDA and the European Medicines Agency use the two-failure version: an inadequate response to at least two antidepressants, despite an adequate trial of each and taking the medicine as prescribed. By that definition at least 30% of people with depression qualify. The same 2023 review points out that a significant share of people called “resistant” are really “pseudo-resistant”: the trials were too short, the dose too low, or the medicine was not taken consistently.2 So the first question in a consultation is not “what’s next?” It is “what has actually been tried, at what dose, for how long, and what happened?”

Staging systems

Clinicians sometimes use staging systems to describe how resistant an episode has been. A 1997 model describes five stages of increasing resistance, with MAOI antidepressants typically reserved for stage III and ECT for stage IV.18 The Maudsley Staging Method (2009) scores three things, the number of treatments tried, how severe the illness is (psychosis scores highest) and how long the episode has lasted, into a total from 3 (minimal resistance) to 15 (severe).19 An international group has proposed a broader idea, “difficult-to-treat depression,” defined as “depression that continues to cause significant burden despite usual treatment efforts.” When remission keeps proving elusive, the first goal shifts toward the best possible symptom control, daily function and quality of life, decided together with the patient. That proposal is expert consensus rather than trial evidence.20

None of these systems tells you what to do. They are a way of making sure the conversation starts from an accurate picture of your history.

The STAR*D numbers, and why two figures are quoted

The best-known large study of medication treatment in steps is STAR*D, published in 2006. It treated 3,671 outpatients with up to four successive steps; 123 people reached the fourth. On the self-report scale the investigators used, remission was 36.8% at step 1, 30.6% at step 2, 13.7% at step 3 and 13.0% at step 4, and the reported cumulative remission rate was 67%. People who needed more steps were more likely to relapse during follow-up.3

In 2023 a different group reanalyzed the patient-level data. They used the blinded Hamilton rating scale that the study protocol had named as its primary outcome, instead of the unblinded clinic-administered scale, and they excluded 99 patients who already scored as remitted when they entered the study and 125 who scored as remitted when they started their next step. With those rules the cumulative remission rate was 35.0%.4 The original investigators’ reply was not reviewed for this page, so the dispute is stated here without taking a side: the two figures come from different rating scales and different rules about who counts. What the original report itself shows, and the reanalysis did not challenge, is that remission became less likely with each medication step after the second; the original investigators also found that people who needed more steps relapsed more often during follow-up. The reanalysis did not examine relapse. That is one reason guidelines consider treatments beyond another medication switch once two adequate trials have failed.3,4,21

How clinicians think about the choice

There is no formula. But the questions a clinician works through are fairly consistent, and it helps to know them in advance, because several of them can only be answered by you.

How severe is this, and is anyone’s safety at risk?

This is the question that can override everything else. When depression comes with psychosis, catatonia, not eating or drinking, or serious thoughts of suicide, the major guidelines agree that ECT moves to the front. The 2022 VA/DoD guideline recommends offering ECT for severe depression with catatonia, psychosis, severe suicidality, or a “need for rapid, definitive treatment response on either medical or psychiatric grounds,” among other listed conditions; it is the only strong recommendation the guideline gives among these treatments.21,22 CANMAT’s 2023 update says that “in the most severe MDEs and/or in life-threatening situations (e.g., severe suicide risk and physical deterioration), electroconvulsive therapy (ECT) should be considered as the first-choice option.”23 The FDA’s 2018 order places ECT devices in class II (cleared with special controls, not “approved”) for treating catatonia or a severe major depressive episode in people age 13 and older “who are treatment-resistant or who require a rapid response due to the severity of their psychiatric or medical condition.”24

How fast do you need to feel better?

Speed differs. In a pooled analysis of five trials (664 people), ketamine reduced symptoms more than ECT 24 hours after the first treatment, but by the end of the course response and remission did not differ.25 In a secondary analysis of the US ELEKT-D trial that was not planned in advance, very severely depressed inpatients improved earlier with ECT than with ketamine, though the two groups were similar by the end of three weeks.12 TMS is given five days a week for four to six weeks before its effect is judged.26,27 Vagus nerve stimulation is the slowest to judge: its trials measure benefit over twelve months and beyond.28,29

What has already been tried, and what happened?

Past response is one of the guides every guideline mentions. All three guidelines say a good response to ECT in the past is a reason to choose ECT again.21,23,30 NICE also says that if ECT did not work well before, a repeat course should only be considered after reviewing whether the earlier course was adequate and considering all other options.30 CANMAT notes that rTMS can be a first choice if it worked in a previous episode or if medication side effects are the main problem.23 In a meta-analysis of nine trials comparing the two, ECT’s advantage over rTMS was concentrated in psychotic depression; in non-psychotic depression rTMS was as effective.8 In the VNS trial, people who had already had ECT or TMS in the current episode did worse overall but gained more from active stimulation than from sham.31

What medical risks apply to you?

ECT requires general anesthesia, and NICE asks clinicians to weigh “the risks associated with a general anaesthetic, any medical comorbidities, potential adverse events, in particular cognitive impairment,” the possible added risk in older people, and “the risks associated with not having ECT.”30 Esketamine carries a boxed warning for sedation, dissociation, respiratory depression, abuse and misuse, and suicidal thoughts in young people, and the label requires blood-pressure checks and at least two hours of observation after each dose.32 VNS requires surgery to implant a stimulator, and shortness of breath was more common with active stimulation in its controlled trial.28 TMS’s most common side effects in its trials were headache and scalp discomfort.26,27

What can your life accommodate?

A treatment you cannot complete is not a treatment. The VA/DoD guideline gives rTMS only a weak recommendation partly because it “requires a considerable time commitment.”21 CANMAT, on the other hand, says that outside emergencies “rTMS should generally be considered before ECT, given its less invasive nature and that individuals may continue to work or attend school while undergoing rTMS.”23 Both are true: TMS takes many visits but does not take you out of your life; ECT takes fewer visits but each one involves anesthesia, recovery time and someone to drive you. The comparison table below gives the schedules.

Which side effects are acceptable to you?

Each treatment has a different profile, and people weigh them differently. For some, any memory effect is a dealbreaker; for others, several weeks of daily visits is. The numbers are in the table below.

What will insurance cover?

Coverage is a real part of the decision. Esketamine must be given in a certified clinic under a federal safety program.32 Intravenous ketamine is FDA-approved as an anesthetic, not for depression, so coverage for depression is not standardized.21 Medicare covers VNS for depression only inside an approved study, and the Medicare criteria require at least four prior failed treatments of adequate dose and duration.33 TMS and ECT have FDA-cleared devices; whether a given plan covers them, and what prior authorization is needed, is something to check before you commit.24,2

What do you prefer?

Preference is not a tiebreaker; it is an input, and there is some evidence it changes outcomes. In a post-hoc survey of 255 people in the ketamine-versus-ECT trial, more participants preferred ketamine; preferring ketamine was associated with responding regardless of which treatment they got; getting the treatment they preferred was associated with response to ketamine (but not ECT) and with fewer side effects in the ECT group. These are associations from a survey done after treatment, not proof that preference causes the outcome.34 CANMAT, NICE and the VA/DoD guideline all frame the final choice as a shared decision.23,30,21

What the guidelines say, in their own words

GuidelineDatePosition
VA/DoD Clinical Practice Guideline for MDD, version 4.021,22February 2022After partial or no response to two or more adequate medication trials, “we suggest offering repetitive transcranial magnetic stimulation” (weak for). “There is insufficient evidence to recommend for or against theta-burst stimulation.” After several adequate medication trials, “we suggest ketamine or esketamine as an option for augmentation” (weak for). ECT is recommended (strong for) for severe depression with catatonia, psychosis, severe suicidality, prior good response, need for rapid definitive response, higher risk from other treatments, or poor response or intolerance to multiple antidepressants. It “suggest[s] against using vagus nerve stimulation outside of a research setting” (weak against).
CANMAT 2023 update23Published 2024ECT “should be considered as the first-choice option” in “the most severe MDEs and/or in life-threatening situations”; elsewhere the guideline says ECT “can be used as first-line treatment in severe illness ... or for patients who had a prior good response to ECT.” “Except in emergent situations ... rTMS should generally be considered before ECT.” “ECT is more efficacious for DTD [difficult-to-treat depression] compared to other noninvasive neuromodulation treatments, although it carries a greater side effect burden.” Quotes ECT response rates “ranging from 65% to 75%.”
NICE NG222, Depression in adults3029 June 2022“Consider electroconvulsive therapy (ECT) for the treatment of severe depression if: the person chooses ECT in preference to other treatments based on their past experience of ECT and what has previously worked for them, or a rapid response is needed (for example, if the depression is life-threatening because the person is not eating or drinking), or other treatments have been unsuccessful.” Clinicians must “reach a shared decision on its use,” make sure consent “is given without pressure or coercion,” and that “the person is aware of their right to change their mind and withdraw consent at any time.” rTMS and VNS are referred to NICE’s separate procedures guidance. NICE’s technology appraisal does not recommend esketamine nasal spray for treatment-resistant depression.
FDA final order on ECT devices24Effective December 26, 2018ECT devices are class II for “catatonia or a severe major depressive episode ... in patients age 13 years and older who are treatment-resistant or who require a rapid response due to the severity of their psychiatric or medical condition.”

These are guidelines, not rules. They differ in emphasis (CANMAT puts rTMS ahead of ECT outside emergencies; the VA/DoD guideline gives ECT its only strong recommendation among these treatments), and none of them was written about you.

The options side by side

The table uses numbers from randomized trials and meta-analyses where they exist, says so when a figure comes from an observational follow-up, and gives the number of people studied, so you can judge how solid each figure is. Response usually means symptoms cut at least in half; remission means close to symptom-free. Different studies use different scales, so numbers are not exactly comparable across rows. Treatment details are on each treatment’s own page.

TreatmentWhat it isTime commitmentEvidence, in numbersMain side effectsFDA and coverage status
TMS A magnetic coil held against the scalp stimulates the left front of the brain while you sit awake in a chair. No anesthesia. Sessions five days a week for four to six weeks in the pivotal trials; a standard session is about 37.5 minutes, and a newer form (theta-burst) takes about 3 minutes and was not worse than the standard session in a 414-person trial (self-rated scalp pain was slightly higher).26,35,27 Against sham: remission 14.1% versus 5.1% in an NIH trial of 190 people (number needed to treat 12); in a network analysis of 113 brain-stimulation trials, standard high-frequency left-sided TMS roughly tripled the odds of response versus sham (odds ratio 3.17, 95% CI 2.29–4.37; an odds ratio of 3 means roughly three times the odds).35,6 Against ECT: remission 33.6% versus 52% (7 trials, 294 people), with ECT’s advantage concentrated in psychotic depression.36,8 In an observational follow-up of 257 patients at 42 practices, 62.5% of the 120 who responded or remitted kept their response for a year; 36.2% of all 257 received a repeat course of TMS during the year.37 Headache (64–65% in a 414-person trial) and scalp discomfort; 4.5% stopped for side effects in the first pivotal trial; in trials against ECT, visual memory and verbal fluency were more affected by ECT than by rTMS.27,26,8 FDA-cleared devices for treatment-resistant depression.2 Weak recommendation in VA/DoD 2022; CANMAT suggests it before ECT outside emergencies.21,23 Plans commonly require prior authorization; coverage rules are plan-specific.
Ketamine (intravenous) A low-dose infusion of an anesthetic drug, given over about 40 minutes while you are monitored.9 Twice a week for three weeks (6 infusions) in the US trial; three times a week, up to 12, in the Swedish trial.9,12,10 Versus ECT in 365 US outpatients and inpatients without psychosis: response 55.4% ketamine versus 41.2% ECT (not worse than ECT).9 Versus ECT in 186 Swedish inpatients: remission 46% ketamine versus 63% ECT (ECT better).10 Pooled across 6 trials (655 people), ketamine was not shown to be non-inferior overall and was inferior in inpatients.11 Faster at 24 hours.25 Among Swedish remitters, 70% relapsed within a year after ketamine and 63% after ECT, not a significant difference.10 Dissociation (feeling detached) during the infusion, blurred or double vision, dizziness; better memory-test scores than ECT at the end of treatment.9,25,38 FDA-approved as an anesthetic, not for depression; used off-label.21 VA/DoD 2022: weak recommendation as an augmentation option.21 Coverage varies.
Esketamine nasal spray A nasal spray form of ketamine (the S-enantiomer), taken in a certified clinic with at least two hours of observation.32,21 Twice a week for weeks 1–4, weekly for weeks 5–8, then every one or two weeks; each visit includes at least two hours of monitoring.32 Versus quetiapine, both added to an antidepressant, in 676 people: remission at 8 weeks 27.1% versus 17.6%; remission with no relapse through 32 weeks 21.7% versus 14.1% (open-label, rater-blinded).39 Boxed warning: sedation, dissociation, respiratory depression, abuse and misuse, suicidal thoughts in people under 25; blood pressure rises; the label says not to drive until the next day after a restful sleep.32 FDA-approved March 5, 2019 with an oral antidepressant for treatment-resistant depression; July 31, 2020 for depressive symptoms with acute suicidal thoughts or behavior (effect on suicide itself not established); January 17, 2025 as monotherapy. Available only through the REMS program.32,21
ECT A brief, controlled seizure produced under general anesthesia with a muscle relaxant, in a hospital procedure area. Three times a week in the US trial (9 treatments over 3 weeks); a median of 6 treatments to remission in the Swedish trial; a driver and recovery time after each treatment.9,12,10 Against sham ECT: effect size −0.91 (6 trials, 256 people); against medication: −0.80 (18 trials, 1,144 people).5 Largest effect of 25 treatments in a network analysis of 69 trials in treatment-resistant depression (odds ratio 12.86 versus placebo or sham, wide interval 4.07–40.63).7 Remission 63% in the Swedish inpatient trial but 22% in the US trial; the trial investigators attribute the gap to differences in patients, ECT protocols and study design.10,40 CANMAT quotes response rates of 65–75%.23 Relapse within a year after successful ECT: in a meta-analysis of 32 studies, 51% of people on continuation medication in the modern-era studies relapsed within a year; 84% within six months on placebo.41,42 Memory: delayed-recall scores fell 9.7 points with ECT versus 0.9 with ketamine at three weeks, with gradual recovery; responders showed no difference from ketamine at 1, 3 and 6 months, but the Swedish trial reported cases of persisting amnesia. Muscle aches, headache, anesthesia risks.9,38,10 FDA class II devices since December 26, 2018 for severe depression that is treatment-resistant or needs a rapid response, age 13 and up.24 Strong recommendation in VA/DoD 2022 for the listed situations; first choice in CANMAT for the most severe or life-threatening episodes.21,23
Vagus nerve stimulation (VNS) A pacemaker-like device implanted surgically in the chest, with a wire to the vagus nerve in the neck, delivering intermittent stimulation for years. Surgery, then ongoing stimulation; the controlled trial measured benefit over 12 months, with further gains in year two.28,29 In 493 people with at least four failed treatments in the current episode (average 13.3 lifetime treatments, 17.8-year episode), the main outcome, time in response on the MADRS scale over months 3–12, did not differ from sham. Secondary measures favored active VNS: clinician-rated benefit 39.6% versus 30.7%, global improvement 53.8% versus 39.8%. Over 80% of those with benefit at 12 months kept it at 18 and 24 months.28,43,29 Shortness of breath more common with active stimulation; the risks of the implant surgery itself.28 FDA-approved for resistant depression in 2005. The 2022 VA/DoD guideline suggests against VNS outside a research setting (weak recommendation, written before the 2025 RECOVER results).21 Medicare covers it only within an approved study, for people with at least four prior failed treatments (NCD 160.18, effective February 15, 2019).33

Reading the ketamine-versus-ECT results

The two big head-to-head trials disagree, and it is worth understanding why rather than picking the one you like. The US trial (ELEKT-D) enrolled people referred for ECT who did not have psychosis, treated most of them as outpatients (88%),12 and found ketamine not worse than ECT; its ECT remission rate was 22%, far below the 63% in the Swedish trial and the response rates guidelines quote.9,40 The Swedish trial (KetECT) enrolled hospital inpatients and found ECT clearly better; its ECT remission rate was 63%.10 Meta-analyses that pool the trials find ECT ahead on average, with the advantage clearest in inpatient trials; ketamine is faster in the first 24 hours, and the most recent pooled analysis found better memory-test scores after ketamine while two earlier ones found no significant cognitive difference (sham means a fake treatment used as a comparison; a meta-analysis pools several trials).44,45,11,25 Within the US trial, the more severely ill inpatients improved earlier with ECT.12 Within the US trial, people with milder depression treated as outpatients did better with ketamine, and very severe inpatients improved faster with ECT.12 The practical reading: if you are severely ill, in hospital, or have psychosis, the evidence leans toward ECT (ketamine has not been tested against ECT in psychotic depression); if you are an outpatient without psychosis, the US trial suggests ketamine is a reasonable option to discuss first; and in both cases the memory trade-off is real but, for responders and on group averages, was not measurable one to six months later.38

Durability is the weak point of all of them

Getting better and staying better are different questions. After successful ECT, about half of people relapse within a year even on continuation medication, and most of that relapse happens in the first six months; without medication, 84% relapsed within six months in a randomized trial.41,42 After ketamine, 70% of Swedish remitters relapsed within a year, against 63% after ECT in the same trial (not a significant difference).10 After TMS, in an observational follow-up, 62.5% of responders held their response for a year, and 36.2% of all patients followed received a repeat course.37 After esketamine, 21.7% of people were in remission at week 8 and still had not relapsed by week 32.39 Whatever we start with, the plan has to include what happens afterward.

Shared decision-making, in practice

“Shared decision-making” gets used a lot in healthcare without much behind it. Here is what it means in this clinic, and what the research says it does.

Two people bring expertise to any medical decision. I bring the medical side: what these treatments do, who they tend to help, what the risks are, what the evidence says and where it runs out. You bring expertise on your own life: what matters to you, what you can tolerate, what your days look like, what you have already tried, what you are afraid of and what you are hoping for. Neither kind of expertise replaces the other. In practice that means I lay out the realistic options, tell you what I would be thinking about if this were my own family member, answer questions as many times as you want to ask them, and we land on a plan we both actually agree with.

What decision aids have been shown to do

A decision aid is a structured tool (cards, a booklet, a website) that lays out the options, their benefits and harms, and helps you sort out what matters to you. The 2024 Cochrane review pooled 209 randomized trials with 107,698 participants, 10 of them in mental health. Compared with usual care, decision aids probably made people more likely to choose the option that matched their own values (relative risk 1.75; moderate-certainty evidence from 21 trials), and with high-certainty evidence they improved knowledge by about 12 points on a 100-point scale, nearly doubled accurate understanding of the risks, reduced the feeling of being uninformed, and reduced the share of people who took a passive role. They did not increase regret about the decision (high certainty). They did not lengthen the consultation when used beforehand, and added about a minute and a half when used during it.13

In depression specifically, a cluster-randomized trial of 117 clinicians and 301 primary-care patients choosing an antidepressant found that a card-based decision aid improved decisional comfort (80% versus 75%), knowledge (65% versus 56%) and involvement (47% versus 33%) without lengthening the visit, but it made no difference to whether people took the medicine or how their depression did.46 A Spanish trial of a web-based aid in 147 primary-care patients likewise improved knowledge and reduced decisional conflict, with no change in what people intended to do.47 A Cochrane review of 15 trials of shared decision-making interventions in mental health found that people may report more shared decision-making right after the encounter (three trials, 534 participants, low-certainty evidence), but whether it improves symptoms is uncertain because that evidence is of very low certainty.48

What has not been shown. We found no randomized trial of a decision aid for choosing among TMS, ketamine, ECT and VNS. One published decision aid built for treatment-resistant depression (developed with industry involvement) was piloted in five patients, where decisional conflict fell from 42.2 to 28.1; it has not been tested in a randomized trial, and its authors point to gaps in the comparative-effectiveness evidence needed to fill it with reliable numbers.14 The decision-aid evidence shows that structured information helps people feel informed and involved. It does not show that it leads to better depression outcomes.46,48

What people who went through it say they needed

Two interview studies of people who chose ECT, one consumer-led with 17 participants and one with 16 patients and relatives, found overlapping themes: information presented calmly and in steps and offered more than once, because people found it hard to retain; clear information about memory effects, which many felt was lacking; family and friends involved in the decision; the chance to hear from people who had the treatment; and support for memory side effects afterward.49,50 Those findings shape how consultations here are run: you can bring anyone you want, you can ask the same question twice, and you will get the memory numbers rather than a reassurance.

Informed consent is a conversation, not a form

Signing a consent form is paperwork. Real informed consent means four things have happened before you sign.

  • You understand the treatment. Well enough to explain to a family member what will happen, what it is designed to do and roughly how it is supposed to work.
  • You understand the realistic range of outcomes, including not responding. The table above gives the response and remission rates and the relapse rates. A consent conversation that gives only the good number is not complete.
  • You understand the alternatives. Including doing nothing different right now, and including treatments not offered here.
  • You are free to say no, to wait, or to change your mind. NICE puts it plainly: consent must be “given without pressure or coercion from the circumstances or clinical setting,” and you must be “aware of [your] right to change [your] mind and withdraw consent at any time.”30

If any of those four pieces is missing when you are asked to sign, the consent is not informed yet, and we keep talking until it is.

What is not approved: psilocybin and MDMA

You will hear about psychedelic treatments, and it is fair to ask about them. Here is the status as of October 26, 2026. More detail is on the psychedelics page.

Psilocybin

Psilocybin is not FDA-approved for depression or any other condition. The two largest peer-reviewed trials are phase 2 trials. In 233 people with treatment-resistant depression, a single 25 mg dose lowered depression scores by 6.6 points more than a 1 mg control dose at three weeks, but the secondary measure of sustained response at 12 weeks did not support the result, 77% of participants had side effects, and suicidal thoughts or behavior occurred in every dose group.51 In 104 people with major depression (not treatment-resistant), a single 25 mg dose beat a niacin placebo by 12.3 points at six weeks, with no serious adverse events but more side effects overall.52 A CANMAT task force (published 2023, evidence to July 2021) graded the evidence as low level, flagged functional unblinding as a major limitation (participants can usually tell whether they got the drug, which undermines the placebo comparison), and called psilocybin experimental, to be used mainly within clinical trials.15 The manufacturer has since reported two larger phase 3 trials by press release (June 23, 2025 and February 17, 2026): a single 25 mg dose beat placebo by 3.6 points at six weeks in 258 people, and two 25 mg doses three weeks apart beat a 1 mg dose by 3.8 points in 581 people, smaller differences than in the phase 2 trial; those results have not been peer-reviewed, and the company says it expects to complete a rolling FDA submission in late 2026.16 Psilocybin is not offered here.

MDMA

MDMA-assisted therapy was studied for post-traumatic stress disorder, not depression. Two phase 3 trials (90 and 104 people) found larger reductions in PTSD symptoms than therapy with placebo, but the FDA committee judged that functional unblinding and selection bias limited what the trials could show.53,54,17 In June 2024 an FDA advisory committee voted 9 to 2 that the data did not show effectiveness and 10 to 1 that the benefits did not outweigh the risks, and on August 9, 2024 the FDA issued a complete response letter declining approval; the letter, made public by the FDA in September 2025, says the sponsor had not shown that the benefit lasts beyond 18 weeks and lists additional safety work needed. The company, renamed Resilient Pharmaceuticals, is reported to have resubmitted the PTSD application in August 2026 without a new phase 3 trial. As of October 26, 2026, MDMA is not FDA-approved for any condition, and no application for depression has been announced.17

If a clinic offers you either of these outside a registered clinical trial, ask what evidence, what oversight and what safety monitoring are behind it.

Questions to bring to the consultation

You do not need answers before we meet. But thinking about these ahead of time usually makes the conversation more useful.

Questions to ask yourself

  • How urgent does this feel to me, honestly? Is anyone worried about my safety?
  • What have I tried, at what dose, for how long, and what did I learn from each attempt?
  • What would I do if I felt better in three months? In six months?
  • Which side effects would be dealbreakers for me: memory, feeling detached for an hour or two, daily visits, anesthesia, surgery?
  • Who in my life do I want involved in this decision, and who can drive me and stay with me if that is needed?
  • What am I most afraid of about treatment? What am I most hoping for?
  • How much certainty do I need before I would be willing to try something?

Questions to ask the clinician

  • By which definition am I “treatment-resistant,” and were my past medication trials actually adequate?1,2
  • What response and remission numbers do you expect for someone like me, and which study do they come from?
  • What is the chance this does not work, and what would we do then?
  • What is the relapse rate after this treatment, and what is the plan to keep me well afterward?41,37,39
  • What are the memory or cognitive effects, measured how, and how long did they last in the trials?9,38,10
  • What will this cost me, what does my insurance require, and is this treatment approved for my situation?
  • Which alternatives are you not offering here, and where would I go for them?
  • If I were your family member, what would you be weighing?

The process, from first call to decision

Every program runs a little differently. This is how it works here, described in general terms; specific scheduling and timing depend on your situation and will be explained to you directly.

  1. Referral. Treatment here starts with a referral from a psychiatric provider. If you already have a psychiatrist, ask them to refer you. The contact details are at the end of this page.
  2. Records. Before the visit we gather what has been tried: medications with doses and durations, psychotherapy, prior neuromodulation, hospitalizations, medical history and current medicines. The more complete this is, the more accurate the conversation about whether past trials were adequate.2
  3. Evaluation. A thorough conversation about your history, what is going on now and what is at stake, together with a medical review of anything that affects anesthesia or other risks. Not a checklist; a real conversation. You are welcome to bring a family member or friend.49,50
  4. Understanding your options. We go through the treatments that might fit: what each does, who it tends to help, the realistic range of outcomes including non-response, the trade-offs, the practical details and the likely cost. I will tell you what I would recommend and why, and why I might be wrong.
  5. A plan we both agree with. If that takes more than one visit, that is fine. Some decisions deserve a second conversation, a second opinion or time with family. Nothing is decided under pressure unless safety truly requires it, and even then the reasons are explained to you.30
  6. Follow-up and adjustment. Whatever we start with is a plan, not a guarantee. We track symptoms with standard questionnaires, ask about side effects rather than assume them, and adjust the plan as we learn. Because relapse is common after ECT, ketamine and TMS, and benefit can fade after any treatment, the plan for staying well is made before the acute treatment ends, not after.41,10,37

Frequently asked

How do I know if neuromodulation is right for me?

We figure it out together. There is no simple rule, but there are patterns: guidelines point to ECT when depression is severe, psychotic, catatonic or dangerous, to rTMS after two or more adequate medication trials and to ketamine or esketamine after several, and to VNS only after many failed treatments.21,23,33 The conversation starts with your history and what you are looking for, and grows from there.

Which treatment works best?

On average, ECT. It has the largest effect against sham treatment and against medication, and it ranks first in network comparisons of brain-stimulation treatments, with wide confidence intervals.5,6,7 It also had more memory side effects than ketamine and than rTMS in the head-to-head trials.9,10,8 “Best on average” is not the same as best for you: for people without psychosis who were referred for ECT, ketamine was no worse than ECT in the largest trial, and in a secondary analysis that was not planned in advance outpatients did somewhat better with ketamine; rTMS matched ECT in non-psychotic depression in a pooled analysis.9,12,8

Do I have to commit to a treatment at my first visit?

No. Good decisions often take more than one conversation. If you need more time, more information or a second opinion, take it. The exception is when safety requires acting quickly, and in that case the reasons will be explained to you and your family.23

What if I am not sure I want to go through with a treatment after we have discussed it?

Then you don’t. You can take time to think, talk with family, come back with more questions, get a second opinion or decide this is not the right time. Guidelines are explicit that consent must be free of pressure and can be withdrawn at any time.30 I would rather you say no to the right treatment at the wrong moment than yes before you are ready.

What happens if the treatment does not work?

Then we talk about what is next. Non-response is common enough with every one of these treatments that we plan for it up front: in the two head-to-head trials above, between 37% (ECT, Swedish inpatients) and 78% (ECT, US outpatients referred for ECT) did not reach remission with ECT, and 54% to 62% did not with ketamine, depending on the trial and setting; 73% did not reach remission at eight weeks with esketamine.10,40,39 Options include adjusting the current treatment, switching to a different one, or rethinking the plan. In the VNS trial, about a third of people with no benefit at 12 months gained a meaningful benefit in the open-label second year.29 It is not a dead end.

What if I start a treatment and want to stop?

You can. I would want to understand what changed, because sometimes it is a side effect we can manage, but you are not locked in. In the trials, 4.5% of people stopped TMS for side effects in the first large sham-controlled trial (301 patients), and in the Swedish ketamine-versus-ECT trial more people stopped ketamine than ECT because of side effects.26,10

Can I bring someone with me?

Yes. Family members, partners and close friends are welcome in the conversation with your consent. People who have been through ECT say involving family and hearing information more than once were among the most helpful things.49,50

What if I want a second opinion?

Get one. For a decision this consequential a second opinion is often a good idea, and I will help you think about where to go for one.

Can a brain scan or EEG tell which treatment will work for me?

Not yet in routine care. Prediction research is active, and CANMAT says plainly that it does not recommend EEG for choosing treatment, and the other guidelines used on this page make no such recommendation.23,21,30 For now the predictors with the most evidence are the clinical ones described above: severity, psychosis, past response and how adequate earlier treatments were.

Get in touch

Consultation

Treatment is started by referral from a psychiatric provider. Please ask your psychiatrist to refer you; referring providers can call the Greenville ECT referral line.

(864) 455-8813

Prisma Health Psychiatry and Behavioral Medicine – Greenville
725 Grove Road, Greenville, SC 29605
Adam Hart, MD · Greenville, SC

Consultations take place at the Grove Road office. If a treatment is chosen, your care team will tell you exactly where and when to come for each step.

Sources

Every number on this page is linked to one of the sources below, numbered in order of first citation. PubMed entries were checked against the NCBI record on October 26, 2026; guideline, regulatory and company sources are dated in the entry.

  1. Gaynes BN, Lux L, Gartlehner G, et al. Defining treatment-resistant depression. Depress Anxiety. 2020;37(2):134-145. PubMed 31638723. Review for CMS and AHRQ: no consensus definition of treatment-resistant depression; the most common definition required at least two prior treatment failures with confirmation of adequate dose and duration; only 17% of intervention studies enrolled samples meeting it.
  2. McIntyre RS, Alsuwaidan M, Baune BT, et al. Treatment-resistant depression: definition, prevalence, detection, management, and investigational interventions. World Psychiatry. 2023;22(3):394-412. PubMed 37713549. The FDA and EMA use inadequate response to a minimum of two antidepressants despite adequate trials and adherence; at least 30% of people with depression meet it; a significant percentage are pseudo-resistant (inadequate trials or non-adherence).
  3. Rush AJ, Trivedi MH, Wisniewski SR, et al. Acute and longer-term outcomes in depressed outpatients requiring one or several treatment steps: a STAR*D report. Am J Psychiatry. 2006;163(11):1905-17. PubMed 17074942. STAR*D: 3,671 outpatients received one step and 123 received four; QIDS-SR16 remission 36.8%, 30.6%, 13.7% and 13.0% at steps 1–4; reported cumulative remission 67%; more steps meant higher relapse in follow-up.
  4. Pigott HE, Kim T, Xu C, et al. What are the treatment remission, response and extent of improvement rates after up to four trials of antidepressant therapies in real-world depressed patients? A reanalysis of the STAR*D study's patient-level data with fidelity to the original research protocol. BMJ Open. 2023;13(7):e063095. PubMed 37491091. Reanalysis of STAR*D patient-level data using the protocol-specified blinded HRSD and excluding 99 patients remitted at entry and 125 remitted when starting the next level: cumulative remission 35.0% versus the reported 67%.
  5. UK ECT Review Group. Efficacy and safety of electroconvulsive therapy in depressive disorders: a systematic review and meta-analysis. Lancet. 2003;361(9360):799-808. PubMed 12642045. Real versus simulated ECT, 6 trials (n=256): standardized effect size −0.91 (95% CI −1.27 to −0.54); ECT versus pharmacotherapy, 18 trials (n=1,144): −0.80 (−1.29 to −0.29).
  6. Mutz J, Vipulananthan V, Carter B, et al. Comparative efficacy and acceptability of non-surgical brain stimulation for the acute treatment of major depressive episodes in adults: systematic review and network meta-analysis. BMJ. 2019;364:l1079. PubMed 30917990. Network meta-analysis, 113 trials (n=6,750): response versus sham, bitemporal ECT OR 8.91 (2.57–30.91), high-dose right unilateral ECT 7.27 (1.90–27.78), intermittent theta-burst 3.20 (1.45–7.08), high-frequency left rTMS 3.17 (2.29–4.37); 83% of trials at low or unclear risk of bias; all treatments at least as acceptable as sham.
  7. Saelens J, Gramser A, Watzal V, et al. Relative effectiveness of antidepressant treatments in treatment-resistant depression: a systematic review and network meta-analysis of randomized controlled trials. Neuropsychopharmacology. 2025;50(6):913-919. PubMed 39739012. Network meta-analysis in treatment-resistant depression (at least two failed trials), 69 RCTs (n=10,285), 25 treatments: six beat placebo or sham (ECT, minocycline, theta-burst, rTMS, ketamine, aripiprazole); OR 1.9 (1.25–2.91) for aripiprazole to 12.86 (4.07–40.63) for ECT; 59% of studies had some risk-of-bias concerns.
  8. Ren J, Li H, Palaniyappan L, et al. Repetitive transcranial magnetic stimulation versus electroconvulsive therapy for major depression: a systematic review and meta-analysis. Prog Neuropsychopharmacol Biol Psychiatry. 2014;51:181-9. PubMed 24556538. Nine trials (n=425): ECT response 64.4% versus 48.7% and remission 52.9% versus 33.6% for high-frequency rTMS; ECT’s advantage mainly in psychotic depression, with rTMS as effective in non-psychotic depression; discontinuation 8.3% versus 9.4%; visual memory and verbal fluency more impaired with ECT in 3 studies.
  9. Anand A, Mathew SJ, Sanacora G, et al. Ketamine versus ECT for Nonpsychotic Treatment-Resistant Major Depression. N Engl J Med. 2023;388(25):2315-2325. PubMed 37224232. ELEKT-D, open-label noninferiority RCT, 5 US sites, nonpsychotic treatment-resistant depression referred for ECT: 403 randomized, 195 treated with ketamine (0.5 mg/kg twice weekly) and 170 with ECT (three times weekly) for 3 weeks; response 55.4% versus 41.2% (difference 14.2 points, 95% CI 3.9–24.2; noninferior); HVLT-R delayed recall change −0.9 versus −9.7 with gradual recovery; quality of life similar; ECT musculoskeletal effects, ketamine dissociation.
  10. Ekstrand J, Fattah C, Persson M, et al. Racemic Ketamine as an Alternative to Electroconvulsive Therapy for Unipolar Depression: A Randomized, Open-Label, Non-Inferiority Trial (KetECT). Int J Neuropsychopharmacol. 2022;25(5):339-349. PubMed 35020871. KetECT, open-label noninferiority RCT, Sweden, 186 inpatients with unipolar depression, ketamine 0.5 mg/kg three times weekly versus ECT, up to 12 sessions: remission 63% ECT versus 46% ketamine (P=.026; difference 95% CI 2–30%); median 6 sessions in both; serious and long-lasting adverse events including persisting amnesia more common with ECT; more dropouts for adverse events with ketamine; relapse within 12 months among remitters 70% ketamine versus 63% ECT (P=.52).
  11. Petrucci ABC, Fernandes JVA, Reis IA, et al. Ketamine versus electroconvulsive therapy for major depressive episode: An updated systematic review and non-inferiority meta-analysis. Psychiatry Res. 2024;339:115994. PubMed 38865906. Meta-analysis including ELEKT-D, 6 RCTs (n=655): ketamine not noninferior to ECT for response overall (risk difference −0.10, −0.26 to 0.05); among inpatients ketamine inferior (−0.15, −0.27 to −0.03); ketamine better post-treatment cognition and less muscle pain, more dissociation.
  12. Jha MK, Wilkinson ST, Krishnan K, et al. Ketamine vs Electroconvulsive Therapy for Treatment-Resistant Depression: A Secondary Analysis of a Randomized Clinical Trial. JAMA Netw Open. 2024;7(6):e2417786. PubMed 38916891. ELEKT-D secondary analysis (n=365; 6 infusions versus 9 ECT treatments over 3 weeks): milder depression (QIDS-SR16 ≤20) and outpatients improved more with ketamine (−7.7 vs −5.6; −8.4 vs −6.2); very severe inpatients improved earlier with ECT (−8.4 vs −6.7) but similarly by the end (−9.0 vs −9.9); not prespecified.
  13. Stacey D, Lewis KB, Smith M, et al. Decision aids for people facing health treatment or screening decisions. Cochrane Database Syst Rev. 2024;1(1):CD001431. PubMed 38284415. Cochrane review, 209 RCTs (107,698 participants; 10 in mental health): decision aids versus usual care increased choices matching personal values (RR 1.75, 1.44–2.13), knowledge (+11.9 points per 100), accurate risk perception (RR 1.94) and participation (passive role RR 0.72), reduced feeling uninformed (−10.0 points), did not change decision regret, and did not lengthen consultations when used beforehand.
  14. Shillington AC, Langenecker SA, Shelton RC, et al. Development of a patient decision aid for treatment resistant depression. J Affect Disord. 2020;275:299-306. PubMed 32734922. Development of a treatment-resistant-depression decision aid to IPDAS standards; pilot in 5 patients: decisional conflict 42.2 to 28.1, decision self-efficacy 86.0 to 95.5; no randomized trial; the authors note gaps in comparative-effectiveness evidence.
  15. Rosenblat JD, Husain MI, Lee Y, et al. The Canadian Network for Mood and Anxiety Treatments (CANMAT) Task Force Report: Serotonergic Psychedelic Treatments for Major Depressive Disorder. Can J Psychiatry. 2023;68(1):5-21. PubMed 35975555. CANMAT task force: only low-level (Level 3–4) evidence for psychedelics in depression; functional unblinding a major limitation; psilocybin considered experimental and recommended primarily within clinical trials.
  16. Compass Pathways press releases (company-reported, not peer-reviewed): COMP005, June 23, 2025, n=258, single 25 mg dose versus placebo, week-6 MADRS difference −3.6 (95% CI −5.7 to −1.5); COMP006, February 17, 2026, n=581, two doses three weeks apart, 25 mg versus 1 mg week-6 difference −3.8 (−5.8 to −1.8); 26-week data July 7, 2026; serious adverse events in the 25 mg arms 5% (COMP005) and 2% (COMP006 Part A); FDA granted a rolling review in April 2026 and the company expects final submission in Q4 2026 (releases of February 17, July 7 and August 5, 2026, read October 26, 2026; excerpt in scouts/raw/links/choose-compass-press-releases-2026-10-26.excerpt.txt). Not FDA-approved as of October 26, 2026.
  17. U.S. Food and Drug Administration, Psychopharmacologic Drugs Advisory Committee, June 4, 2024, final summary minutes (NDA 215455, midomafetamine capsules, Lykos Therapeutics, PTSD): effectiveness vote Yes 2, No 9; benefit-risk with REMS Yes 1, No 10 (fda.gov). Complete response letter, NDA 215455, issued August 9, 2024 (company and press reports) and released publicly by the FDA on September 4, 2025: durability not shown beyond 18 weeks; additional safety characterization and bias-minimization required. MAPS press release, August 10, 2026 (GlobeNewswire): Resilient Pharmaceuticals (formerly Lykos) reported to have resubmitted the PTSD application without a new phase 3 trial; no FDA acceptance or decision found. Not FDA-approved as of October 26, 2026; no depression application announced. Copies in scouts/raw/links/choose-*.
  18. Thase ME, Rush AJ. When at first you don't succeed: sequential strategies for antidepressant nonresponders. J Clin Psychiatry. 1997;58 Suppl 13:23-9. PubMed 9402916. Five-stage model of treatment resistance; MAOIs (with or without lithium) typically reserved for stage III and ECT for stage IV.
  19. Fekadu A, Wooderson S, Donaldson C, et al. A multidimensional tool to quantify treatment resistance in depression: the Maudsley staging method. J Clin Psychiatry. 2009;70(2):177-84. PubMed 19192471. Maudsley Staging Method: treatment (5-point), severity (1 = subsyndromal to 5 = severe with psychosis) and episode duration (3-point); total 3 (minimal) to 15 (severe); reasonable face and predictive validity.
  20. McAllister-Williams RH, Arango C, Blier P, et al. The identification, assessment and management of difficult-to-treat depression: An international consensus statement. J Affect Disord. 2020;267:264-282. PubMed 32217227. International consensus defining difficult-to-treat depression as “depression that continues to cause significant burden despite usual treatment efforts”; shifts the first goal from remission to optimal symptom control, function and quality of life through shared decision-making; based largely on expert consensus.
  21. U.S. Department of Veterans Affairs and Department of Defense. VA/DoD Clinical Practice Guideline for the Management of Major Depressive Disorder, Version 4.0, February 2022 (healthquality.va.gov; evidence reviewed through January 2021). Recommendation 17 (rTMS after two or more adequate pharmacologic trials, Weak for), 18 (theta-burst stimulation, neither for nor against), 19 (ketamine or esketamine augmentation after several adequate pharmacologic trials, Weak for), 20 (ECT for severe MDD with listed conditions, Strong for), 33 (suggests against vagus nerve stimulation outside a research setting, Weak against). Discussion notes that rTMS “requires a considerable time commitment,” that ketamine is FDA-approved for general anesthesia, and that esketamine’s effectiveness in preventing suicide has not been established.
  22. McQuaid JR, Buelt A, Capaldi V, et al. The Management of Major Depressive Disorder: Synopsis of the 2022 U.S. Department of Veterans Affairs and U.S. Department of Defense Clinical Practice Guideline. Ann Intern Med. 2022;175(10):1440-1451. PubMed 36122380. Synopsis of the February 2022 VA/DoD guideline: 36 recommendations developed with GRADE.
  23. Lam RW, Kennedy SH, Adams C, et al. Canadian Network for Mood and Anxiety Treatments (CANMAT) 2023 Update on Clinical Guidelines for Management of Major Depressive Disorder in Adults: Réseau canadien pour les traitements de l'humeur et de l'anxiété (CANMAT) 2023 : Mise à jour des lignes directrices cliniques pour la prise en charge du trouble dépressif majeur chez les adultes. Can J Psychiatry. 2024;69(9):641-687. PubMed 38711351. CANMAT 2023 update (published 2024): ECT “should be considered as the first-choice option” in the most severe or life-threatening episodes; “except in emergent situations ... rTMS should generally be considered before ECT”; ECT response rates “ranging from 65% to 75%”; “ECT is more efficacious for DTD compared to other noninvasive neuromodulation treatments, although it carries a greater side effect burden”; emphasizes shared decision-making.
  24. U.S. Food and Drug Administration. Neurological Devices; Reclassification of Electroconvulsive Therapy Devices. Final order, Federal Register document 2018-27809 (federalregister.gov), effective December 26, 2018: class II for “catatonia or a severe major depressive episode (MDE) associated with major depressive disorder (MDD) or bipolar disorder (BPD) in patients age 13 years and older who are treatment-resistant or who require a rapid response due to the severity of their psychiatric or medical condition.”
  25. Shi ZM, Lan XJ, Chen Q, et al. Intravenous ketamine versus electroconvulsive therapy for major depressive disorder or bipolar depression: A meta-analysis of randomized controlled trials. J Affect Disord. 2025;371:45-53. PubMed 39549887. Meta-analysis, 5 RCTs (n=664): 24 hours after the first treatment ketamine reduced symptoms more (SMD −0.49, −0.96 to −0.03); at the end of treatment response (RR 0.96) and remission (RR 0.91) did not differ; ketamine more dissociation, blurred vision, dizziness and double vision; ECT more muscle pain.
  26. O'Reardon JP, Solvason HB, Janicak PG, et al. Efficacy and safety of transcranial magnetic stimulation in the acute treatment of major depression: a multisite randomized controlled trial. Biol Psychiatry. 2007;62(11):1208-16. PubMed 17573044. Multisite sham-controlled RCT, 301 medication-free patients: sessions five times per week for 4–6 weeks; active TMS superior at weeks 4 and 6; remission about twice as frequent at week 6; 4.5% dropped out for adverse events, mostly transient scalp discomfort or pain.
  27. Blumberger DM, Vila-Rodriguez F, Thorpe KE, et al. Effectiveness of theta burst versus high-frequency repetitive transcranial magnetic stimulation in patients with depression (THREE-D): a randomised non-inferiority trial. Lancet. 2018;391(10131):1683-1692. PubMed 29726344. THREE-D noninferiority RCT, 414 patients with treatment-resistant depression: 3-minute theta-burst versus 37.5-minute 10 Hz sessions, 5 days a week for 4–6 weeks; HRSD-17 fell from 23.5 to 13.4 and 23.6 to 13.4; headache in 64% and 65%; dropouts 6% and 8%.
  28. Conway CR, Aaronson ST, Sackeim HA, et al. Vagus nerve stimulation in treatment-resistant depression: A one-year, randomized, sham-controlled trial. Brain Stimul. 2025;18(3):676-689. PubMed 39706521. RECOVER, 12-month double-blind sham-controlled RCT, 493 adults with marked treatment-resistant depression: the primary outcome (percent time in MADRS response, months 3–12) did not separate active from sham; clinician (CGI-I, P=0.004) and patient (QIDS-SR, P=0.049) ratings favored active VNS; shortness of breath more common with active VNS; 88.4% completed.
  29. Conway CR, Rush AJ, Aaronson ST, et al. Durability of the benefit of vagus nerve stimulation in markedly treatment-resistant major depression: a RECOVER trial report. Int J Neuropsychopharmacol. 2026;29(1):pyaf080. PubMed 41529263. RECOVER open-label extension (n=214): about 79% of those with substantial benefit at 12 months kept it at 18 and 24 months; 30.6% and 37.8% of those without meaningful benefit at 12 months gained it by 18 and 24 months.
  30. National Institute for Health and Care Excellence. Depression in adults: treatment and management (NG222), published 29 June 2022, recommendations 1.13.1–1.13.8 (ECT; consent wording in 1.13.3), 1.14.1 (rTMS, cross-reference to interventional procedures guidance) and 1.15.1–1.15.2 (VNS cross-reference; esketamine nasal spray not recommended in NICE technology appraisal guidance). nice.org.uk/guidance/ng222, checked October 26, 2026.
  31. Aaronson ST, Conway CR, Gordon C, et al. Prognostic and Prescriptive Predictors of Treatment Response to Adjunctive VNS Therapy in Major Depressive Disorder: A RECOVER Trial Report. J Clin Psychiatry. 2025;86(3):25m15850. PubMed 40673907. RECOVER predictors: a history of ECT or of TMS in the current episode predicted poorer outcomes overall but greater benefit from active versus sham VNS.
  32. U.S. Food and Drug Administration. SPRAVATO (esketamine) nasal spray prescribing information, NDA 211243, revised 04/2025 (Medication Guide revised 09/2026; REMS modified through June 2026; Drugs@FDA); Drugs@FDA approval history: original approval March 5, 2019; depressive symptoms in MDD with acute suicidal ideation or behavior July 31, 2020; monotherapy for treatment-resistant depression January 17, 2025. Boxed warning: sedation, dissociation, respiratory depression, abuse and misuse, suicidal thoughts and behaviors; monitor for at least two hours after each dose; available only through the SPRAVATO REMS; dosing twice weekly in weeks 1–4, weekly in weeks 5–8, then every one or two weeks.
  33. Centers for Medicare & Medicaid Services. National Coverage Determination 160.18, Vagus Nerve Stimulation, version 3, effective February 15, 2019, implementation July 22, 2020 (cms.gov): FDA approved VNS for resistant depression in 2005; Medicare covers VNS for treatment-resistant depression only within a CMS-approved Coverage with Evidence Development trial, for patients with at least four prior failed treatments of adequate dose and duration; non-covered outside such a study.
  34. Sanacora G, Barnett BS, Hu B, et al. Patient preference effects in a randomized comparative effectiveness study of electroconvulsive therapy and ketamine for treatment resistant depression: An ELEKT-D trial secondary analysis. Psychiatry Res. 2025;347:116411. PubMed 40049091. ELEKT-D post-hoc preference survey (n=255): more participants preferred ketamine; preferring ketamine was associated with response regardless of assignment; receiving the preferred treatment was associated with response to ketamine but not ECT, and with fewer adverse events in ECT.
  35. George MS, Lisanby SH, Avery D, et al. Daily left prefrontal transcranial magnetic stimulation therapy for major depressive disorder: a sham-controlled randomized trial. Arch Gen Psychiatry. 2010;67(5):507-16. PubMed 20439832. NIH-sponsored sham-controlled RCT, 190 analyzed: 37.5-minute daily weekday sessions for 3 weeks, continued up to 3 more weeks in improvers; remission 14.1% active versus 5.1% sham (OR 4.2, NNT 12); about 30% remitted in open-label follow-up; most remitters had low treatment resistance.
  36. Berlim MT, Van den Eynde F, Daskalakis ZJ. Efficacy and acceptability of high frequency repetitive transcranial magnetic stimulation (rTMS) versus electroconvulsive therapy (ECT) for major depression: a systematic review and meta-analysis of randomized trials. Depress Anxiety. 2013;30(7):614-23. PubMed 23349112. Seven RCTs (n=294), high-frequency rTMS versus ECT: after an average of 15.2 rTMS and 8.2 ECT sessions, remission 33.6% (38/113) versus 52% (53/102), OR 0.46, NNT 6 favoring ECT; dropouts did not differ.
  37. Dunner DL, Aaronson ST, Sackeim HA, et al. A multisite, naturalistic, observational study of transcranial magnetic stimulation for patients with pharmacoresistant major depressive disorder: durability of benefit over a 1-year follow-up period. J Clin Psychiatry. 2014;75(12):1394-401. PubMed 25271871. Naturalistic 12-month follow-up of 257 patients at 42 practices: of 120 who responded or remitted with acute TMS, 75 (62.5%) kept their response through 12 months; 36.2% received repeat TMS (mean 16.2 treatment days).
  38. Kumpf KT, Wilkinson ST, Hu B, et al. Comparing the Cognitive Effects of Repeated Intravenous Ketamine and Electroconvulsive Therapy in Patients With Treatment-Resistant Depression: A Secondary Analysis of the ELEKT-D Trial. J Clin Psychiatry. 2025;86(4):25m15781. PubMed 40900112. ELEKT-D cognition (n=365): ECT recipients scored worse on all four cognitive tasks at the end of treatment (P<.001); among responders no group differences at 1, 3 and 6 months.
  39. Reif A, Bitter I, Buyze J, et al. Esketamine Nasal Spray versus Quetiapine for Treatment-Resistant Depression. N Engl J Med. 2023;389(14):1298-1309. PubMed 37792613. ESCAPE-TRD, open-label rater-blinded phase 3b RCT: esketamine (n=336) versus extended-release quetiapine (n=340), both added to an SSRI or SNRI; remission at week 8 27.1% versus 17.6% (P=0.003); remission at week 8 with no relapse through week 32 21.7% versus 14.1%.
  40. Ekstrand J, Takamiya A, Nordenskjold A, et al. Ketamine or ECT? What Have We Learned From the KetECT and ELEKT-D Trials?. Int J Neuropsychopharmacol. 2024;27(1):pyad065. PubMed 38114073. Commentary on the two trials: ECT remission 63% in KetECT versus 22% in ELEKT-D; ketamine remission 46% versus 38%.
  41. Jelovac A, Kolshus E, McLoughlin DM. Relapse following successful electroconvulsive therapy for major depression: a meta-analysis. Neuropsychopharmacology. 2013;38(12):2467-74. PubMed 23774532. Meta-analysis of 32 studies: after successful ECT, 51.1% (95% CI 44.7–57.4) relapsed within 12 months on continuation medication, 37.7% within 6 months; continuation ECT 6-month relapse 37.2%; antidepressants halved relapse versus placebo (RR 0.49, NNT 3.3).
  42. Sackeim HA, Haskett RF, Mulsant BH, et al. Continuation pharmacotherapy in the prevention of relapse following electroconvulsive therapy: a randomized controlled trial. JAMA. 2001;285(10):1299-307. PubMed 11255384. RCT, 84 ECT remitters, 24 weeks: relapse 84% on placebo, 60% on nortriptyline, 39% on nortriptyline plus lithium.
  43. Conway CR, Sackeim HA, Rush AJ, et al. The RECOVER Trial of Vagus Nerve Stimulation in Markedly Treatment-Resistant Depression: Critical Findings, Lessons Learned, and Future Directions. Am J Psychiatry. 2026;183(10):708-720. PubMed 42642808. RECOVER summary: at least four failed trials in the current episode, an average of 13.3 lifetime antidepressant treatments and a 17.8-year current episode; QIDS-C benefit 39.6% versus 30.7%, CGI-I 53.8% versus 39.8%, function 43.3% versus 37.5%, quality of life 45.4% versus 37.1%; over 80% of those with benefit at 12 months retained it at 18 and 24 months.
  44. Rhee TG, Shim SR, Forester BP, et al. Efficacy and Safety of Ketamine vs Electroconvulsive Therapy Among Patients With Major Depressive Episode: A Systematic Review and Meta-analysis. JAMA Psychiatry. 2022;79(12):1162-1172. PubMed 36260324. Meta-analysis, 6 trials (n=340), all inpatient settings: ECT versus ketamine SMD −0.69 (−0.89 to −0.48) favoring ECT; no significant differences in cognition or serious adverse events; low-to-moderate quality.
  45. Menon V, Varadharajan N, Faheem A, et al. Ketamine vs Electroconvulsive Therapy for Major Depressive Episode: A Systematic Review and Meta-analysis. JAMA Psychiatry. 2023;80(6):639-642. PubMed 37043224. Meta-analysis, 5 RCTs (141 ketamine, 137 ECT): SMD −0.39 (−0.81 to 0.02); in the two stronger trials −0.45 (−0.75 to −0.14); response RR 1.27 (1.06–1.53) and remission RR 1.43 (1.12–1.82) favoring ECT; all trials at high risk of bias.
  46. LeBlanc A, Herrin J, Williams MD, et al. Shared Decision Making for Antidepressants in Primary Care: A Cluster Randomized Trial. JAMA Intern Med. 2015;175(11):1761-70. PubMed 26414670. Cluster RCT, 117 clinicians and 301 primary-care patients choosing an antidepressant with a decision aid: decisional comfort 80% versus 75%, knowledge 65% versus 56%, involvement 47% versus 33%; no difference in visit length, adherence or depression scores.
  47. Perestelo-Perez L, Rivero-Santana A, Sanchez-Afonso JA, et al. Effectiveness of a decision aid for patients with depression: A randomized controlled trial. Health Expect. 2017;20(5):1096-1105. PubMed 28295915. RCT, web decision aid (68 versus 79 primary-care patients with depression): better knowledge and less decisional conflict (both P<.001); no change in treatment intention or preference for participation.
  48. Aoki Y, Yaju Y, Utsumi T, et al. Shared decision-making interventions for people with mental health conditions. Cochrane Database Syst Rev. 2022;11(11):CD007297. PubMed 36367232. Cochrane review, 15 RCTs of shared decision-making interventions in mental health: effects on clinical outcomes uncertain (very low certainty); user-reported shared decision-making improved immediately after the encounter (SMD 0.63, 0.26–1.01; 3 studies, 534 participants; low certainty).
  49. Wells K, Scanlan JN, Gomez L, et al. Decision making and support available to individuals considering and undertaking electroconvulsive therapy (ECT): a qualitative, consumer-led study. BMC Psychiatry. 2018;18(1):236. PubMed 30041602. Consumer-led interviews with 17 people who had ECT: more information, offered repeatedly; involve family and friends; hear from people who have had ECT; more support for memory side effects.
  50. Rovers J, Schönberger L, Loef D, et al. Exploring the Decision-Making Process for Electroconvulsive Therapy in Patients With Major Depressive Disorder and Their Relatives After Treatment. J ECT. 2025;41(3):197-205. PubMed 39998858. Interviews with 16 patients and relatives who chose ECT: information presented calmly and stepwise; gaps about memory loss and difficulty retaining information; support from family, clinicians and peers helped; stigma and fear hindered.
  51. Goodwin GM, Aaronson ST, Alvarez O, et al. Single-Dose Psilocybin for a Treatment-Resistant Episode of Major Depression. N Engl J Med. 2022;387(18):1637-1648. PubMed 36322843. Phase 2 trial in treatment-resistant depression, single dose of psilocybin 25 mg (n=79), 10 mg (75) or 1 mg (79): week-3 MADRS change −12.0, −7.9 and −5.4; 25 mg versus 1 mg −6.6 (−10.2 to −2.9); sustained response at 12 weeks not supported; adverse events in 77%; suicidal ideation or behavior in all groups.
  52. Raison CL, Sanacora G, Woolley J, et al. Single-Dose Psilocybin Treatment for Major Depressive Disorder: A Randomized Clinical Trial. JAMA. 2023;330(9):843-853. PubMed 37651119. Phase 2 trial in major depression, psilocybin 25 mg (n=51) versus niacin (n=53): day-43 MADRS difference −12.3 (−17.5 to −7.2); no serious adverse events; higher rates of overall and severe adverse events with psilocybin.
  53. Mitchell JM, Bogenschutz M, Lilienstein A, et al. MDMA-assisted therapy for severe PTSD: a randomized, double-blind, placebo-controlled phase 3 study. Nat Med. 2021;27(6):1025-1033. PubMed 33972795. Phase 3 trial of MDMA-assisted therapy for severe PTSD (n=90): CAPS-5 change −24.4 versus −13.9 with placebo plus therapy (d=0.91).
  54. Mitchell JM, Ot'alora G M, van der Kolk B, et al. MDMA-assisted therapy for moderate to severe PTSD: a randomized, placebo-controlled phase 3 trial. Nat Med. 2023;29(10):2473-2480. PubMed 37709999. Second phase 3 trial for moderate-to-severe PTSD (n=104): CAPS-5 change −23.7 versus −14.8 (d=0.7); severe adverse events in 5 versus 2 participants.

Last reviewed: October 26, 2026 · Adam Hart, MD · Greenville, SC. This page is educational and does not substitute for a conversation with your clinician.

Important

The information on this page is for educational purposes only. It is not medical advice and is not a substitute for evaluation, diagnosis, or treatment by a qualified healthcare professional. No physician-patient relationship is established through this website. If you are experiencing a medical or psychiatric emergency, call 911 or go to the nearest emergency room. If you are having thoughts of suicide or self-harm, call or text 988 to reach the Suicide and Crisis Lifeline.