Treatment-resistant depression (TRD) usually means major depression that has not improved enough after at least two antidepressant trials, each taken at an adequate dose, for an adequate length of time, and actually taken as prescribed. That is the definition the FDA and the European Medicines Agency use. There is no single agreed definition beyond that, and the details matter: a surprising share of "resistant" depression turns out to be under-treated, misdiagnosed or complicated by another medical problem.
Having TRD is not a verdict that nothing will work. It means the first, simplest steps have not been enough, and that it is time for a more careful review and a wider set of options, which this guide walks through.
The common definition, and why it is fuzzy
A 2020 review prepared for the federal Centers for Medicare & Medicaid Services found no consensus definition of TRD. The most common one required at least two prior treatment failures with confirmed adequate dose and duration, but experts did not agree on what "adequate" means, and only 17% of treatment studies enrolled patients who met that most common definition.
Researchers have built more detailed staging systems to capture this. The Maudsley Staging Model, for example, scores how many treatments have failed, how long the episode has lasted and how severe it is, rather than giving a simple yes or no. A related idea, "difficult-to-treat depression", shifts the goal from full remission alone toward symptom control, getting back to work and life, and quality of life.
Not better, partly better, or well?
Definitions also blur the line between "no response" and "partial response". In research, response usually means a drop of at least 50% on a depression rating scale, and remission means symptoms have fallen to a low, near-normal range. Many people sit in between: noticeably better, but still flat, tired, foggy or anxious. The World Psychiatry review points out that even among STAR*D participants classed as "in remission", only 10% were completely free of symptoms, and that people who remit but still report poor quality of life are at greater risk of relapse. So if you are functioning but not yourself, that is worth raising, even if a chart says you "responded".
Insurers use their own versions of the definition, which is why the number of medications you need to have tried before TMS is covered varies by plan. We cover that in detail in how many antidepressants you have to try before TMS is covered.
What makes a medication trial "adequate"
Before anyone calls depression treatment-resistant, a psychiatrist should check whether each past trial really had a fair chance. Four things count.
| Question | What it means in practice |
|---|---|
| Was it the right diagnosis? | Major depression, not bipolar depression, a medical cause or another condition driving the symptoms |
| Was the dose adequate? | At least the dose recommended for depression in the drug's labeling, not a starter dose that was never raised |
| Was it long enough? | Generally 4 to 6 weeks at an optimal dose; some people need longer |
| Was it taken consistently? | Missed doses, stopping early because of side effects, or on-and-off use all weaken a trial |
On duration, a 2023 review in World Psychiatry puts the usual adequate trial at 4 to 6 weeks at optimal dosing, while noting that 60% of people who reached remission on the first STAR*D medication did so after week 6. The same review summarizes data suggesting that about 20% of people who have not responded by week 4 respond in weeks 5 to 8, and about 10% in weeks 9 to 12. That is why "it did nothing after two weeks" is not, by itself, a failed trial.
A helpful exercise before your appointment is to write down every antidepressant you have taken: the name, highest dose, how long you took it, why you stopped, and what changed. Some clinics use a structured form for this, such as the Antidepressant Treatment History Form.
Pseudo-resistance: when it only looks treatment-resistant
"Pseudo-resistance" describes depression that seems resistant but is really the result of something fixable. The main culprits are worth checking one by one.
Under-dosing, short trials and missed doses
Non-adherence is common and often not anyone's fault: side effects, cost, stigma and simple forgetting all play a part. Using a common definition of less than 80% adherence to the prescribed regimen, about 30% to 50% of people prescribed antidepressants are non-adherent in the first, acute phase of treatment. If you skipped doses or stopped because of side effects, say so; it changes the plan, and it is not a failure on your part.
A different diagnosis, especially bipolar disorder
Inaccurate diagnosis is one of the most common reasons for apparent resistance; the World Psychiatry review estimates that about half of people with major depression are not correctly diagnosed. A classic pattern is someone who has tried several antidepressants in a row without benefit and actually has bipolar disorder, in which depression is often the first episode to appear. Ask your psychiatrist about screening for past high-energy or "wired" periods; this guide to bipolar disorder vs ordinary mood swings explains what to look for.
Medical conditions that drive or mimic depression
Other conditions should be identified and treated, including substance and alcohol use disorders, anxiety disorders, personality disorders and some physical illnesses such as an underactive thyroid (hypothyroidism). Untreated sleep apnea is another to consider: in a small 2010 study of 17 people with major depression that persisted despite medication and who also had sleep apnea, Beck Depression Inventory scores fell from 19.7 to 10.8 after 2 months of CPAP treatment. That study had no control group, so it shows a possibility to check, not a guaranteed fix. The physician's guide on drlewis.com covers the full medical workup before calling depression treatment-resistant.
How your body processes the medication
Some people break down certain antidepressants unusually fast because of variations in liver enzymes called CYP2D6 and CYP2C19, which can leave blood levels too low to work. The World Psychiatry review says pharmacogenetic testing cannot be recommended routinely in TRD but may be warranted in select situations. The site's article on reasons to consider genetic testing in psychiatry discusses when it can help.
How common is treatment-resistant depression?
Estimates depend heavily on the definition. TRD is often said to affect about 30% of people receiving antidepressants in research settings, while real-world estimates range from 6% to 55%. A 2021 U.S. analysis estimated that of 8.9 million adults with medication-treated major depression in a year, 2.8 million (30.9%) had TRD.
The large NIMH-funded STAR*D study is often used to show how the odds change with each step. It reported remission rates of 36.8%, 30.6%, 13.7% and 13.0% for the first to fourth treatment steps and a cumulative remission rate of 67%. A 2023 reanalysis that followed the original protocol more strictly put cumulative remission at 35.0%, and the original investigators disputed that reanalysis. Either way, the chance of remission with each additional medication step fell after the first two, which is one reason guidelines bring in other kinds of treatment at that point. For a closer look at the study and its limits, see the drlewis.com article after two antidepressants: what STAR*D tells us.
Your options when depression is treatment-resistant
There is no single correct next step. The list below groups the main options; your psychiatrist will weigh them against your history, preferences, side effects and access.
| Option | What it involves | Where it stands |
|---|---|---|
| Optimizing the current medication | Raising the dose or allowing more time | Worth checking first if the trial was short or under-dosed |
| Switching antidepressants | Moving to a different drug or class | Evidence on whether switching helps in TRD is mixed |
| Combining or augmenting | Adding a second antidepressant, certain antipsychotics, lithium or thyroid hormone | Some antipsychotic add-ons are proven in partial responders |
| Adding psychotherapy | CBT or another structured therapy alongside medication | Offers significant relief when added to antidepressants |
| TMS | Daily magnetic stimulation sessions over several weeks | First-line for TRD in the CANMAT 2023 guideline |
| Esketamine or ketamine | Supervised nasal spray (Spravato) or off-label IV ketamine | Established as efficacious in TRD; second-line in CANMAT |
| ECT | Brief seizures induced under general anesthesia, usually 6 to 12 sessions | First-line for severe depression; response rates 65% to 75% |
| Vagus nerve stimulation | Surgically implanted stimulator | Third-line in CANMAT |
The table draws on the 2023 World Psychiatry review and CANMAT 2023. The review rates the evidence for extending, switching and combining antidepressants as mixed, notes that psychotherapies are not established as stand-alone treatments in TRD but add meaningful relief to medication, and calls IV ketamine and intranasal esketamine established treatments. For a deeper comparison of each choice, read TMS or another antidepressant? and TMS vs ketamine therapy. The site also has an older piece on augmentation strategies for treatment-resistant depression.
Where TMS fits
TMS was developed for exactly this situation. Its FDA-cleared indication is major depression in adults who have not improved enough with prior antidepressant medication in the current episode. CANMAT 2023 rates rTMS as first line for treatment-resistant depression, with the strongest level of evidence for short-term (acute) benefit, and the U.S. VA/DoD guideline suggests offering it after partial or no response to two or more adequate medication trials. CANMAT also says rTMS should generally be considered before ECT, except in emergencies, because it is less invasive.
TMS is not a cure, and it does not work for everyone. In sham-controlled trials, the difference in remission between real and sham TMS is typically about 9 to 18 percentage points, and patients with fewer prior medication failures tended to do better in the NIMH trial. The TMS candidate checklist can help you see whether it is worth raising with your psychiatrist.
When TMS is not the right next step
Some situations call for a different first move. Expert consensus recommends that people with psychotic symptoms or acute suicidal thinking be considered for treatments with established efficacy in those situations, such as ECT, rather than starting with TMS. If bipolar disorder is suspected, the diagnosis and mood-stabilizing treatment come first, because the FDA guidance lists bipolar disorder among the groups with a lack of evidence for TMS efficacy or safety.
If you are thinking about suicide or are in crisis, call or text 988 (the 988 Suicide & Crisis Lifeline) or go to the nearest emergency room.
What to do next
- Gather your medication history: names, highest doses, durations, side effects and reasons for stopping.
- Ask for a review of the diagnosis, including screening for bipolar disorder, thyroid problems, sleep apnea and substance use.
- Start tracking your symptoms with a validated scale such as the PHQ-9, so progress is measured rather than guessed; these signs your treatment is not working show what to watch for.
- Talk through the full option list above, including TMS, with your psychiatrist.
You can find more in the TMS guide, read about TMS treatment for depression, or see the site's overview of treatment-resistant depression care.
Sources
- Treatment-resistant depression: definition, prevalence, detection, management, and investigational interventions. World Psychiatry, 2023. pmc.ncbi.nlm.nih.gov
- Defining treatment-resistant depression. Depression and Anxiety, 2020. pubmed.ncbi.nlm.nih.gov
- The Prevalence and National Burden of Treatment-Resistant Depression and Major Depressive Disorder in the United States. Journal of Clinical Psychiatry, 2021. pubmed.ncbi.nlm.nih.gov
- Effect of CPAP treatment on residual depressive symptoms in patients with major depression and coexisting sleep apnea: Contribution of daytime sleepiness to residual depressive symptoms. Sleep Medicine, 2010. pubmed.ncbi.nlm.nih.gov
- Acute and longer-term outcomes in depressed outpatients requiring one or several treatment steps: a STAR*D report. American Journal of Psychiatry, 2006. pubmed.ncbi.nlm.nih.gov
- A reanalysis of the STAR*D study's patient-level data with fidelity to the original research protocol. BMJ Open, 2023. pubmed.ncbi.nlm.nih.gov
- Canadian Network for Mood and Anxiety Treatments (CANMAT) 2023 Update on Clinical Guidelines for Management of Major Depressive Disorder in Adults. Canadian Journal of Psychiatry, 2024. pmc.ncbi.nlm.nih.gov
- VA/DoD Clinical Practice Guideline for the Management of Major Depressive Disorder. U.S. Department of Veterans Affairs / Department of Defense, 2022. healthquality.va.gov
- 510(k) K160703 (NeuroStar TMS Therapy System). U.S. FDA, 2016. accessdata.fda.gov
- Daily left prefrontal transcranial magnetic stimulation therapy for major depressive disorder: a sham-controlled randomized trial. Archives of General Psychiatry, 2010. pubmed.ncbi.nlm.nih.gov
- Efficacy and safety of deep transcranial magnetic stimulation for major depression: a prospective multicenter randomized controlled trial. World Psychiatry, 2015. pmc.ncbi.nlm.nih.gov
- Consensus Recommendations for the Clinical Application of Repetitive Transcranial Magnetic Stimulation (rTMS) in the Treatment of Depression. Journal of Clinical Psychiatry, 2018. pmc.ncbi.nlm.nih.gov
- Class II Special Controls Guidance Document: Repetitive Transcranial Magnetic Stimulation (rTMS) Systems. U.S. FDA, 2011. fda.gov
- Transcranial Magnetic Stimulation (LCD L33398). CMS Medicare Coverage Database, 2026. cms.gov
FAQs
Is treatment-resistant depression a separate diagnosis?
No. It describes how major depression has responded to treatment so far, not a different illness. That is why definitions vary between researchers, regulators and insurers.
Does treatment-resistant mean my depression is permanent?
No. It means the first steps have not been enough. Many people improve with a different approach, such as an add-on medication, therapy, TMS, esketamine or ECT.
Do therapy trials count toward treatment resistance?
Most definitions count only medication trials, which researchers see as a limitation. Some insurers, including Medicare in New York, also require a trial of evidence-based psychotherapy before they cover TMS.
If I stopped an antidepressant because of side effects, does that count?
It can. Several insurers count documented intolerance as a failed trial for TMS coverage, but a trial stopped after a few days is not the same as one that did not work at a full dose. See stopped antidepressants because of side effects?
Should I get genetic testing before trying another medication?
Not routinely. Current reviews say pharmacogenetic testing may help in select cases, for example when someone has had unusual side effects at low doses or no effect at high doses. Ask your psychiatrist whether your history fits.
