Most people starting an antidepressant are told to wait four to six weeks before expecting any improvement, but research shows that timeline is incomplete and sometimes discouraging. Many people notice early changes in sleep, energy, or concentration within the first one to two weeks, even though the fuller picture of improvement takes longer to develop. Understanding what happens week by week helps you know what to look for, when to be patient, and when a medication genuinely isn't working.
The standard advice (give it four to six weeks) is not wrong, but it is better understood as "when to judge the full effect" rather than "when anything starts to happen at all." A medication that does nothing in the first several weeks is unlikely to work by week eight, and recognizing early signs of response gives you and your prescriber useful information rather than wasted time.
What happens in the first week on an antidepressant?
The earliest effects of an SSRI or SNRI are not usually mood changes. You might notice shifts in sleep quality, appetite, or energy level before you feel meaningfully less depressed or anxious, though some people do report mood improvement within days, and that is not placebo.
A systematic review analyzing data from more than six thousand patients found that treatment with SSRIs was associated with measurable clinical improvement by the end of the first week, with a mean reduction in depression rating scale scores significantly greater than placebo (Taylor et al., 2006). The difference was modest (roughly one point on a multi-point scale), but it was consistent and it predicted later response. In practical terms, that might look like sleeping more soundly for a few nights in a row, or noticing food tastes more normal, or feeling slightly less physically heavy, before the underlying depression itself has shifted.
The research also shows that one-third of the total symptom improvement people experience during six weeks of SSRI treatment is observed within the first week (Mihilli et al., 2026). That is a striking finding, because it contradicts the common clinical message that "nothing will happen for at least two weeks." Something does happen for many people, and noticing it matters.
What those early changes are not, however, is the full answer. Week one is not when you will know whether the medication is working; it is when you will know whether it is doing anything at all.
Do sleep, appetite, and energy improve before mood?
This is one of the most repeated pieces of guidance in psychiatry, and the evidence for it is weaker than the repetition suggests.
SSRIs act on serotonin systems that regulate sleep, appetite, and energy, so the logic is straightforward: those neurovegetative symptoms should respond first, with mood and cognitive symptoms following weeks later. In practice, the pattern is less clear-cut. Some studies have found that vegetative symptoms improve before mood, while others have found they improve together (Bhar et al., 2008).
What can be said with more confidence is that early changes in sleep and appetite do not guarantee mood improvement, and early mood improvement can happen without clear vegetative changes. Both patterns occur. The usefulness of watching for neurovegetative improvement is not that it proves the medication will work, but that it gives you something concrete to measure during a period when subjective mood is still unreliable.
The timeline differs by symptom. Appetite and sleep disturbances may shift within one to two weeks. Energy and concentration tend to lag (they often improve last, not first), which is one reason people describe feeling "less depressed but still exhausted" in the early weeks. That is not a stalled response; it is an incomplete one, and waiting another few weeks is how you find out whether the rest follows.
What should you expect by week two?
If an antidepressant is going to work for you, week two is when you are likely to start recognizing that something is different. The changes may still be subtle (sleeping an extra hour without waking, eating a meal without forcing it, or going an afternoon without intrusive negative thoughts), but they are more consistent than the fleeting shifts of week one.
This is also the week when side effects, if they are going to be a problem, are usually most noticeable. Nausea, headache, jitteriness, or sexual side effects often appear within the first few days and then either settle or persist into week two. If side effects are severe enough that you are considering stopping the medication, week two is when to contact your prescriber rather than waiting it out or quitting on your own. Some side effects fade; others do not, and switching medications early is often the right call rather than enduring weeks of nausea in the hope that it will eventually be worth it.
The research shows that the greatest rate of change on depression rating scales occurs during the first two weeks, then plateaus around weeks six to eight (Mihilli et al., 2026). That does not mean improvement stops after two weeks; it means the steepest part of the slope is early, and the later weeks are a slower climb rather than a sudden breakthrough.
How long until you feel "better," not just "less bad"?
For most people, the shift from "symptoms are quieter" to "I feel like myself again" takes four to eight weeks. That timeline aligns with the clinical trial standards: the FDA typically requires antidepressant trials to run at least six to eight weeks, because that is how long it takes for a meaningful proportion of participants to reach response or remission.
In the STAR*D trial (the largest real-world study of depression treatment, following nearly three thousand adults treated with citalopram), remission rates were twenty-eight percent at twelve weeks, and a substantial portion of those who responded did so after six weeks of treatment rather than before (Trivedi et al., 2006). That means many people who eventually benefited from the medication would have abandoned it if the trial had ended at four weeks.
The practical implication is that week four is not the endpoint; it is the midpoint. If you have noticed any early improvement by week two or three, continuing to week six or eight is how you find out whether that early signal develops into a fuller response. If you have noticed nothing by week four, that is useful information but not yet a final answer, and your prescriber will likely recommend continuing to at least week six before making a decision.
When do you know it's not working?
The absence of any improvement by four to six weeks is a strong signal that the medication, at that dose, is unlikely to work. One analysis found that patients who showed no onset of response by four to six weeks had a seventy-three to eighty-eight percent chance of still having no response by week eight (Mihilli et al., 2026).
That does not mean the medication should be stopped at week six; it means the dose should be reconsidered, or the medication switched, or augmentation discussed. "Not working" is not the same as "nothing will work," and the STAR*D data make this clear: roughly half of people did not respond to their first antidepressant, and many of those people responded to a second or third agent (Trivedi et al., 2006).
What constitutes "not working" is also not always obvious. Partial improvement (sleeping better and feeling less physically heavy, but still anhedonic and unable to concentrate) is a response, not a failure, and it is information your prescriber uses to decide whether to increase the dose, add a second agent, or switch. Stopping a medication because it "only helped a little" can be the wrong move if the right move is adjusting it.
What happens if you stop before giving it enough time?
The most common way a trial of an antidepressant fails is not that the medication does not work; it is that the trial was never completed. Stopping at two weeks because you expected more, or at four weeks because a friend said theirs worked faster, abandons the trial before the timeline research predicts a response.
The other common pattern is stopping because of side effects that would have resolved if given another week. Nausea, headache, and mild jitteriness from an SSRI often peak in the first week and fade by week two or three. Sexual side effects and persistent insomnia tend not to fade, and those are worth discussing with your prescriber early rather than waiting them out.
The risk of stopping too soon is not just that you miss a medication that would have worked; it is that you are left with no information about whether it worked, which makes the next decision harder. A medication taken for ten days and stopped because of nausea has not been tested. One taken for eight weeks without improvement has been, and the next step is clearer.
How do you track whether it's actually working?
Memory for how you felt four weeks ago is unreliable, and improvement is often gradual enough that you do not notice it from the inside. The alternative is measurement-based care: completing a brief, structured symptom rating scale every week or two, and using the results during appointments to make decisions.
Research on this approach is unambiguous. A review of fifty-one studies found that virtually all trials with frequent, timely feedback of patient-reported symptoms to the provider during medication management significantly improved outcomes, while infrequent assessment and one-time screening did not (Fortney et al., 2017). The difference is not what scale you use; it is that you use one consistently, and that the prescriber sees the results and acts on them.
Practically, that means a good follow-up visit should answer "is this better than last month, and by how much?" with something more durable than a shared impression. If your prescriber is not using a rating scale, you can track symptoms yourself with tools like the PHQ-9 for depression or GAD-7 for anxiety, both freely available and validated, and bring the results to your visit.
What if you tried an antidepressant before and it didn't work?
That history is one of the most useful things you can bring to an evaluation, as long as it is specific. What was tried, at what dose, for how long, and what exactly happened: no effect, partial effect, or side effects that ended the trial before it was complete?
A medication stopped after two weeks because of nausea has not actually been tested. One taken for three months at a low dose without change is a different situation, and the next step is different too: possibly a higher dose of the same medication, possibly a switch to a different class, possibly augmentation with a second agent.
"Nothing works for me" frequently turns out to mean "nothing has yet been given a fair trial with monitoring." That distinction changes everything, because the problem is not that you are treatment-resistant; the problem is that treatment has not yet been delivered in a way that gives it a fair chance.
Getting started
Laura Gabriel, PMHNP-BC, provides psychiatric medication management for adults by secure telehealth video to people located in Connecticut, Massachusetts, New Hampshire, Maine, Rhode Island, Virginia, and Kentucky. The initial evaluation clarifies the diagnosis, reviews what has been tried before, and builds a treatment plan that may or may not include medication. If it does, the follow-up visits track response and adjust the plan based on what is and is not changing. You can read more about starting Lexapro or starting Zoloft for drug-specific timelines, or send a message to get started. Initial evaluations are typically $250–$350 before insurance reimbursement, and follow-up visits are $125–$200. Coverage details are on the insurance page.
This post is general information, not medical advice. Never start, stop, or change a prescribed medication without speaking to your prescriber.
References
Bhar, S. S., Gelfand, L. A., Schmid, S. P., Gallop, R., DeRubeis, R. J., Hollon, S. D., Amsterdam, J. D., Shelton, R. C., & Beck, A. T. (2008). Sequence of improvement in depressive symptoms across cognitive therapy and pharmacotherapy. Journal of Affective Disorders, 110(1-2), 161–166. https://pmc.ncbi.nlm.nih.gov/articles/PMC2566850/
Fortney, J. C., Unützer, J., Wrenn, G., Pyne, J. M., Smith, G. R., Schoenbaum, M., & Harbin, H. T. (2017). A tipping point for measurement-based care. Psychiatric Services, 68(2), 179–188. https://pubmed.ncbi.nlm.nih.gov/27582237/
Mihilli, S., Levitt, A., Solmi, M., Sinyor, M., & Schaffer, A. (2026). When do antidepressants "kick in": Addressing the 2–4-week myth. Canadian Journal of Psychiatry. Advance online publication. https://pmc.ncbi.nlm.nih.gov/articles/PMC13152993/
Taylor, M. J., Freemantle, N., Geddes, J. R., & Bhagwagar, Z. (2006). Early onset of selective serotonin reuptake inhibitor antidepressant action: Systematic review and meta-analysis. Archives of General Psychiatry, 63(11), 1217–1223. https://pmc.ncbi.nlm.nih.gov/articles/PMC2211759/
Trivedi, M. H., Rush, A. J., Wisniewski, S. R., Nierenberg, A. A., Warden, D., Ritz, L., Norquist, G., Howland, R. H., Lebowitz, B., McGrath, P. J., Shores-Wilson, K., Biggs, M. M., Balasubramani, G. K., Fava, M., & STARD Study Team. (2006). Evaluation of outcomes with citalopram for depression using measurement-based care in STARD: Implications for clinical practice. The American Journal of Psychiatry, 163(1), 28–40. https://pubmed.ncbi.nlm.nih.gov/16390886/