Treatment Resistant Depression Explained: What the Term Means and What Gets Looked at Next

Reviewed and updated on August 8, 2026.

This article is educational and independent. It is not medical advice, a diagnosis, or a treatment recommendation, and it is not a substitute for care from a qualified professional. Only a licensed clinician who has evaluated you can advise on your situation. Treatment approaches and availability vary by provider and by state.

The phrase treatment resistant depression usually arrives in a room where somebody has already been trying for a long time. Two medications, maybe three. Months of waiting to feel different. A doctor writes something in the chart, uses the term out loud, and it lands as a verdict about you rather than a description of what the treatments have done so far.

It isn’t a verdict. It’s a category clinicians use to mark a point in a sequence: the standard first steps didn’t produce enough improvement, so the plan changes. The word “resistant” describes the depression’s response to specific interventions that have been tried. It says nothing about effort, willingness, or character, and it isn’t a permanent label attached to a person.

What this article covers: how the term is generally defined and why the definition is looser than it sounds, what an “adequate trial” actually means and why that detail decides whether the label applies at all, what a careful clinician re-examines before adding anything new, and the general categories of next steps that exist. All of it is educational. None of it is a recommendation, and nothing here can tell you what belongs in your own plan.

What treatment resistant depression actually means

The most commonly used definition is an inadequate response to two or more antidepressant trials of adequate dose and adequate duration, given for a current episode of major depressive disorder. The National Institute of Mental Health and the broader research literature generally use some version of that two-trial threshold.

Some version, though. There is no single agreed definition, and that matters more than it looks. Research groups differ on whether the two trials must be from different medication classes, on whether a course of psychotherapy counts as a trial, on what counts as “inadequate” response, and on how partial improvement should be scored. Some frameworks use staged models that rank severity by how many and what kinds of treatments haven’t worked. Some clinicians prefer terms like “difficult-to-treat depression” precisely because “resistant” carries a tone nobody intends.

So when two clinicians use the phrase, they may mean slightly different things. Asking what someone means by it is a legitimate question, not a challenge.

A related distinction that gets blurred in ordinary conversation:

Terms that sound similar and mean different things

Term	General meaning	What it does not mean
No response	Little or no improvement in symptoms after an adequate trial	That nothing else exists to try
Partial response	Measurable improvement, but symptoms still clearly present and disruptive	Treatment failure; this often changes the plan rather than restarting it
Remission	Symptoms reduced to minimal or absent for a sustained period	A guarantee the episode will not return
Relapse	Symptoms returning after a period of improvement	That the treatment never worked
Treatment resistant depression	Inadequate response to two or more adequate antidepressant trials in the current episode	A permanent condition, or a statement about the person
Difficult-to-treat depression	A broader framing emphasizing ongoing management rather than a fixed threshold	A formal diagnostic code

Why “adequate trial” is the phrase that decides everything

Two things have to be true before a medication trial counts: enough of it, for long enough. Both get missed constantly, and when either one is missing, the depression hasn’t actually been shown to resist anything.

Duration is the more common problem. Antidepressants generally take several weeks before a meaningful change in mood shows up, and the professional literature typically treats something in the range of six to eight weeks at a therapeutic dose as the minimum window for judging response. Plenty of people stop at week three because nothing happened. From a clinical standpoint, that’s not a completed trial. It’s an interrupted one.

Dose is the other. A medication kept at a starting level that was never adjusted upward hasn’t been tested at a treatment level. Only the prescriber can make that judgment, and this article cannot and does not give any guidance on dose. The relevant point for a reader is simply that “I took it and it didn’t work” and “I completed an adequate trial” are different statements, and a good history-taking conversation will try to sort out which one applied to each medication in your past.

A third factor sits alongside both: whether the medication was actually taken consistently. This gets asked about a lot, and it can feel accusatory. It generally isn’t meant that way. Side effects, cost, pharmacy gaps, and simply forgetting are all ordinary reasons a trial ends up incomplete, and none of them reflect badly on anyone. The reason it’s asked is that adding a new treatment on top of an untested one produces a mess nobody can interpret later.

Practical thing worth doing: write down your own medication history before an appointment. Name, roughly when you took it, roughly how long, and what happened. Most people can’t reconstruct four years of prescriptions on the spot, and the reconstruction is genuinely useful clinical information.

What a careful clinician re-examines first

Before anything gets added, the usual move is to go back over the fundamentals. This part frustrates people who came in wanting a new option and instead got more questions. There’s a reason for it: a meaningful share of apparent non-response turns out to be something other than a depression that resists antidepressants.

The areas typically revisited include:

• Diagnostic accuracy. Whether the picture fits major depressive disorder or something that overlaps with it. This is the single most consequential re-check.
• Bipolar spectrum features. Periods of elevated, expansive, or unusually energized mood are easy to miss in a history focused on low periods, partly because people rarely come to an appointment to report the weeks they felt great. Depression occurring within bipolar disorder is generally managed differently, and this is a standard thing to screen for before labeling non-response.
• Medical contributors. Thyroid function, anemia, vitamin deficiencies, and other general medical conditions can produce or worsen depressive symptoms. Bloodwork often gets ordered at this stage.
• Sleep. Untreated sleep apnea and chronic insomnia both interact heavily with mood, and treating them can change the picture on their own.
• Substance use. Alcohol in particular. It’s a depressant, it disrupts sleep architecture, and it interacts with treatment in ways that are easy to underestimate.
• Adherence and pharmacy history. Covered above, and usually reviewed with the prescription record rather than memory alone.
• Co-occurring conditions. Anxiety disorders, post-traumatic stress, obsessive-compulsive disorder, ADHD, and personality-related difficulties can all shape response, and an untreated co-occurring disorder can hold symptoms in place.
• Psychosocial load. An unsafe housing situation, an abusive relationship, chronic pain, caregiving with no relief. Medication doesn’t remove a stressor that’s still running.

Reviewing all of that takes time. It may take more than one appointment, and it may involve a referral for a fuller psychiatric evaluation. That is not a stall. It’s the step that prevents years of adding things on top of an unexamined foundation.

The general categories of next steps clinicians consider

What follows is educational description of what exists, presented so the terms are recognizable when a clinician uses them. It is not a menu, not a ranking, and not a suggestion that any of it applies to any particular person. Suitability depends on diagnosis, medical history, prior response, and a clinical evaluation that no article can perform.

General categories of options discussed in this situation

Category	What it generally involves	Typical setting	General state of the evidence
Switching	Changing to a different antidepressant, sometimes in a different class	Outpatient, prescriber-managed	Well established as a standard next step; response rates decline modestly with each successive trial
Augmentation	Adding a second agent alongside the existing one to enhance response	Outpatient, prescriber-managed	Several strategies have trial support; the specific choice is a clinical decision with its own monitoring requirements
Psychotherapy combined with medication	Structured therapy such as cognitive behavioral therapy running alongside medication	Outpatient, weekly	Combination generally shows advantages over either alone for many people; often underused at this stage
Transcranial magnetic stimulation (TMS)	Non-invasive magnetic pulses delivered to a targeted brain region; a course typically runs daily on weekdays for several weeks	Outpatient clinic, awake, no anesthesia	Cleared for use after inadequate response to medication; evidence supports benefit for a meaningful subset
Esketamine and ketamine-related treatments	Rapid-acting approaches administered under supervision with a monitoring period afterward	Certified clinic settings with observation requirements	Evidence supports short-term effects for some people; longer-term data and maintenance questions are less settled
Electroconvulsive therapy (ECT)	Brief electrical stimulation delivered under general anesthesia to induce a controlled seizure, given as a series of sessions	Hospital or specialized outpatient suite	Among the more effective options for severe depression; carries cognitive side effect considerations that require informed discussion
Higher levels of care	Day treatment or intensive outpatient programs providing frequent monitoring while medication is adjusted	Structured program, living at home	Useful for stabilization and for close observation during changes

A few notes on the entries people ask about most.

TMS stands for transcranial magnetic stimulation. Sessions are relatively brief, you’re awake and can drive yourself home, and the schedule is the demanding part: most protocols run every weekday for four to six weeks. People generally describe the sensation as a tapping on the scalp.

ECT stands for electroconvulsive therapy, and public perception of it is largely shaped by films made decades ago. Modern practice uses general anesthesia and a muscle relaxant, and the procedure itself is brief. Memory effects, particularly around the treatment period, are a genuine and well-documented consideration that gets discussed as part of consent. It remains one of the more effective options for severe depression, and it is generally considered in specific clinical situations rather than routinely.

Ketamine-related treatments are the area where marketing has outrun the evidence most visibly. Supervised, regulated administration of an approved product in a certified setting is a different thing from a clinic offering infusions with limited oversight, and the two get discussed as if they were the same. Any conversation about these belongs with a prescriber who knows your history.

Psychotherapy deserves its own mention because it gets skipped. When several medication trials haven’t produced enough change, adding structured therapy is a standard consideration rather than an afterthought, and it’s one of the few categories where the addition doesn’t introduce new medication interactions to manage.

What the research generally shows, and where it’s thinner than the marketing

Large sequenced-treatment research in depression has generally found a consistent pattern: a meaningful proportion of people reach remission with a first antidepressant, a smaller proportion with the second, and progressively smaller proportions with each subsequent step. Response rates decline as you move down the sequence. That’s the honest shape of the data, and it’s also the reason the diagnostic re-check earlier in this article matters so much.

Alongside that, several things hold up reasonably well in the literature:

• Continuing to change the plan is generally better than staying on something that isn’t working, even though the odds per step get narrower
• Combining medication with structured psychotherapy generally performs better than either alone for many people
• Systematic measurement of symptoms over time, rather than relying on impression, is associated with better outcomes because it catches partial response that would otherwise be missed
• Treating co-occurring conditions and sleep problems can change response to depression treatment

Where the evidence is thinner than the promotional language around it:

• Long-term maintenance data for the newer rapid-acting approaches, including how long benefits persist and what ongoing treatment should look like
• Predicting in advance who will respond to which option; genetic testing marketed for this purpose has not been shown to reliably guide medication selection, and professional bodies have generally been cautious about it
• Head-to-head comparisons between the major non-medication options, which are scarce
• Outcomes for people with complex medical histories or multiple co-occurring conditions, who are frequently excluded from trials
• Supplements and devices sold directly to consumers for depression, where claims routinely exceed the evidence

Be skeptical of any clinic or product advertising a specific success percentage without saying what was measured, in whom, and over what period. Reputable programs describe what they track and are willing to discuss who doesn’t respond.

The part nobody writes about: what several failed trials feels like

There’s a specific kind of tiredness that comes from starting a new medication for the fourth time. You know the routine. Six weeks of waiting, side effects in the first two, hope you try not to have because you’ve had it before. Then the appointment where nothing much has changed.

People often describe two reactions at once. Relief that the difficulty has a name and is recognized, and a sinking sense that being categorized this way means the options are running out.

The first reaction is reasonable. The second is worth examining, because the label describes a history of responses, not a ceiling. Options later in a sequence are different in kind, not just more of the same, and some of them have their strongest evidence precisely in people for whom earlier steps didn’t work. That’s not a promise of any particular outcome for any particular person. It’s a correction to the assumption that the list has ended.

Two practical things that people in this situation often find useful: keeping a simple written record of what’s been tried and what happened, and bringing one other person to appointments when possible. Not to speak for you. To remember what was said.

If the weight of this gets heavy, that’s something to say out loud to a clinician rather than carry between appointments. And the crisis resources at the top of this page are free, confidential, and available at any hour.

What the label does not mean

It doesn’t mean nothing will help. It means specific medications, in a specific sequence, didn’t produce enough improvement.

It doesn’t mean the diagnosis is confirmed. Non-response is one of the more common reasons a diagnosis gets revisited, and revisiting it is a standard part of the process rather than a sign the earlier clinician was careless.

It doesn’t mean medication has failed permanently. Response can change with dose adjustments, with a different agent, with the treatment of a co-occurring condition, or with time.

It doesn’t mean you did something wrong. Depression that responds slowly or partially is common enough to have a name and a research literature.

And it doesn’t say anything about what your insurance will authorize for a next step. Prior authorization, coverage rules for procedures, and appeals are a separate topic entirely, handled on our sister site at lawyers.kalmausam.in.

Questions worth bringing to a prescriber

Write these down and take them in. Appointments are short, and the useful questions are the ones that get asked before the conversation runs out of time.

• What do you mean by treatment resistant depression in my case, and which definition are you using?
• Looking at my history, which past medication trials do you consider adequate in dose and duration, and which don’t count?
• Has my diagnosis been reviewed recently, and is there anything that should be re-evaluated?
• Have we checked the medical contributors, including thyroid and other bloodwork?
• Have we looked at sleep, including whether sleep apnea has ever been assessed?
• Is there a co-occurring condition that might be holding symptoms in place?
• What are you considering as the next general step, and what’s the reasoning behind that rather than an alternative?
• How will we measure whether it’s working, and at what point will we decide?
• What side effects would you want me to report, and how do I reach you between appointments?
• Would adding structured psychotherapy be worth considering alongside whatever we do?
• Is a referral for a second opinion or a fuller psychiatric evaluation reasonable at this point?
• If the next step doesn’t help, what would you look at after that?

The last question is often the most valuable. Knowing that a clinician has a plan beyond the immediate step changes how the next six weeks feel.

How this fits with levels of care

Medication changes at this stage sometimes happen alongside more structured treatment, because being seen daily or several times a week makes it easier to catch what’s happening. A partial hospitalization program or an intensive outpatient program can provide that monitoring while someone continues living at home.

Therapy also gets added or restarted at this point for many people. Our explainers on cognitive behavioral therapy and dialectical behavior therapy describe what those approaches involve in practice, and our guide to provider types and credentials explains who can prescribe, who can’t, and what the various licenses mean.

Frequently asked questions

Is treatment resistant depression an official diagnosis?

Not in the way major depressive disorder is. It’s a descriptive clinical term with several competing definitions rather than a standalone diagnostic category, which is why asking a clinician what they mean by it is a fair question.

How many medications have to be tried before the term applies?

Most definitions use two or more adequate trials during the current episode. Whether the trials must be from different classes, and whether psychotherapy counts, varies between frameworks.

What counts as an adequate trial?

Generally a therapeutic dose maintained long enough to judge response, often described as six to eight weeks in the professional literature. A medication stopped after two weeks, or never adjusted from a starting level, typically isn’t counted as a completed trial.

Why does it take so long to know if a medication is working?

The changes involved unfold over weeks rather than days. Sleep and appetite sometimes shift earlier than mood, which is why prescribers ask about those separately rather than only asking whether you feel better.

Could the diagnosis be wrong?

It’s one of the first things clinicians re-examine, and it’s a common enough finding that the re-check is standard practice. Bipolar spectrum features, medical conditions, and untreated co-occurring disorders are all part of that review.

Does this mean I’ll need ECT or TMS?

No. Those are two of several general categories that exist, and whether either is appropriate for any individual depends on a clinical evaluation. Many people’s plans change in other ways entirely, and no article can tell you what belongs in yours.

Is therapy still worth it if medication hasn’t worked?

Combining structured psychotherapy with medication generally shows advantages over either alone for many people, and it’s frequently underused at this stage. Whether and which type is a conversation with your clinician.

Can genetic testing tell me which medication will work?

The tests marketed for this haven’t been shown to reliably guide medication selection, and professional organizations have generally urged caution about the claims made for them. Some clinicians use results as one small input among many.

Does alcohol really make a difference?

It’s a depressant, it fragments sleep, and it interacts with treatment. Clinicians ask about it because it can meaningfully affect response, not to make a moral point about drinking.

What if my current prescriber is out of ideas?

Asking for a referral for a second opinion or for a consultation with someone who focuses on complex mood disorders is a normal request. Most clinicians take it as reasonable rather than as a rejection.

Do symptoms ever improve without another medication change?

They can. Treating a sleep disorder, addressing a thyroid problem, resolving a major stressor, or adding therapy sometimes shifts the picture. That’s exactly why the re-examination step exists before anything new gets added.

How should I keep track of everything I’ve tried?

A single page works: medication name, approximate start and stop dates, roughly how long you took it, whether the dose changed, side effects, and whether anything improved. Bring the same page to every appointment and update it.

Final thoughts

If treatment resistant depression is the phrase being used about your care, the one concrete step worth taking before the next appointment is building that written history of what’s been tried. It takes twenty minutes and a pharmacy printout, and it makes the difference between a conversation based on memory and one based on the actual record. Take it in, and start by asking which of those past trials your clinician counts as adequate.

Sources

• National Institute of Mental Health (NIMH), depression
• Substance Abuse and Mental Health Services Administration (SAMHSA)
• National Institutes of Health / National Library of Medicine
• U.S. Food and Drug Administration, treatment authorizations and safety information

This article is for general informational and educational purposes only. It does not constitute medical advice, a diagnosis, a treatment recommendation, or a substitute for evaluation and care by a qualified health professional. Descriptions of conditions, therapies, procedures, and medications are general and educational; individual experience, suitability, risks, and outcomes vary substantially, and treatment practices change over time. This site is independently operated. It is not a healthcare provider, a treatment facility, a licensed clinician, or a government agency, and it cannot evaluate, diagnose, or treat anyone. Never start, stop, or change any medication or treatment based on anything you read here. Always consult a licensed clinician about your own care, and if you are in crisis, use the free resources listed above.