Antidepressants work better than placebo for most people with depression, but the size of that benefit depends heavily on how severe your symptoms are. Across all 22 commonly prescribed antidepressants, the average effect size over placebo is modest, with a standardized difference of about 0.3 on clinical scales. That number, however, masks a wide range: people with very severe depression see substantial benefits, while those with mild symptoms may get little more than what a placebo provides.
How Much Better Than Placebo?
Every approved antidepressant performs better than a sugar pill in large trials, but “better” needs context. The overall drug-placebo difference of 0.3 on standardized scales is considered a small to modest effect. When patients rate their own symptoms rather than having a clinician score them, the difference shrinks further. To put this in perspective, many common medications in other areas of medicine produce effect sizes above 0.8.
That said, response rates tell a more encouraging story. In large meta-analyses, roughly 48% to 63% of people taking antidepressants achieve a meaningful response, defined as at least a 50% reduction in symptoms. The catch is that placebo groups also improve substantially, often in the range of 30% to 40%, which is why the net advantage of the drug itself can appear small in statistical terms. The placebo response in depression trials is unusually high compared to most medical conditions, partly because regular clinic visits, structured assessments, and the expectation of improvement all have real therapeutic effects.
Severity Changes the Picture Dramatically
A landmark patient-level meta-analysis published in JAMA found that the drug-placebo gap widens as depression gets worse. For people with mild to moderate symptoms, the effect size was just 0.11, which is statistically negligible. For severe depression, it rose to 0.17, still below the threshold for a small effect. But for very severe depression, the effect size jumped to 0.47, a meaningful clinical difference.
Another way to grasp this: the number needed to treat (how many people need to take the medication for one additional person to benefit beyond placebo) was 16 for mild-to-moderate depression, 11 for severe, and just 4 for very severe. In practical terms, if you have very severe depression, there’s roughly a 1-in-4 chance the medication will help you in a way that a placebo would not. If your depression is mild, that number drops to about 1 in 16.
This is why the American College of Physicians recommends cognitive behavioral therapy alone as the initial treatment for mild depression, while suggesting antidepressants or therapy (or both) for moderate to severe cases.
SSRIs vs. Older Antidepressants
The two most commonly compared classes are SSRIs (the newer, widely prescribed type) and tricyclics (an older class). In head-to-head trials, they perform nearly identically when you count every patient who started treatment: about 48% responded to SSRIs and 48.6% to tricyclics. Among patients who actually completed their full course of treatment, tricyclics had a slight edge, with response rates of 68% versus 63%.
The real difference is in tolerability. About 22% of people on tricyclics dropped out due to side effects, compared to 16% on SSRIs. This matters because a medication only works if you keep taking it. SSRIs cause fewer problems with sedation, dry mouth, and cardiovascular effects, which is why they became the standard first-line choice despite being no more effective on paper.
When You Can Expect to Feel a Difference
The conventional wisdom that antidepressants take four to six weeks to kick in is outdated. A meta-analysis of 76 placebo-controlled trials found that 60% of the total improvement seen at six weeks was already apparent by week two. One-third of the full effect showed up in the first week alone. The therapeutic response is actually greatest in the earliest days of treatment, with smaller incremental gains each week after that.
This doesn’t mean you’ll feel completely better in a week. It means the trajectory becomes visible early. If you’ve noticed zero change after two to three weeks at an adequate dose, that’s a signal worth discussing with your prescriber rather than waiting another month.
What Happens if the First Medication Doesn’t Work
About 31% of people treated with antidepressants meet the criteria for treatment-resistant depression, typically defined as failing to respond to two or more adequate medication trials. That’s roughly 2.8 million adults in the United States at any given time. This is not rare, and it doesn’t mean nothing will help. It usually means the treatment strategy needs to change, whether that’s switching medication classes, adding a second medication, or incorporating therapy.
Adding Therapy Makes a Measurable Difference
Combining antidepressants with structured psychotherapy consistently outperforms either treatment alone. In one major trial of chronic depression, the combination produced a 48% remission rate, compared to 29% for medication alone and 33% for therapy alone. That’s a substantial jump.
The benefit of adding therapy is especially pronounced for people with more severe symptoms. In one study, patients with greater symptom severity who received both cognitive therapy and medication achieved a 73% recovery rate, compared to 54% for medication alone. For people with chronic depression who weren’t responding well to standard treatment, adding group-based therapy to ongoing care increased remission from 6% to 26%.
Therapy also helps after you’ve gotten better. A meta-analysis of 13 studies found that adding psychotherapy for patients who responded to antidepressants reduced the risk of relapse by 22 percentage points. That protective effect is one of the strongest arguments for combination treatment: medication can get you into remission, and therapy can help you stay there.
Staying on Medication vs. Stopping
A large trial published in the New England Journal of Medicine tracked patients in primary care who had been stable on antidepressants for at least a year. Half continued their medication and half tapered off under medical supervision. By one year, 39% of those who stayed on medication had relapsed, compared to 56% of those who stopped. Continuing treatment roughly halved the risk of relapse over that period.
Stopping antidepressants also comes with a physical adjustment. About 35% of people experience discontinuation symptoms when they stop, compared to 14% in placebo groups. Common symptoms include dizziness, electric shock-like sensations, nausea, insomnia, vivid dreams, and irritability. These are usually mild to moderate and temporary, but they can be uncomfortable enough to make people think their depression is returning, which complicates the decision to stop.
Putting the Numbers Together
Antidepressants are a genuinely effective treatment for moderate to severe depression, with the benefit growing larger as symptoms get worse. For mild depression, the evidence supports trying therapy first. For everyone else, the choice between medication, therapy, or both is more about personal preference and access than about one being clearly superior. The clinical guidelines reflect this: cognitive behavioral therapy and antidepressants perform about equally as standalone treatments for moderate to severe depression, with no significant difference in response or remission rates after 8 to 16 weeks. Combination treatment may offer an edge, particularly for chronic or severe cases, though the additional benefit over medication alone varies by study.
The most important factor in whether antidepressants “work” may be persistence. Nearly a third of people don’t respond to their first or second try, but that doesn’t mean the third option won’t help. The medications are imperfect tools with real but modest average effects, and their value depends enormously on matching the right treatment to the right person at the right severity level.