AI-researched · adversarially reviewed · human-checked · corrections public
BEFORE YOU READ — HOW DO YOU RULE?
The Verdict
For severe depression, antidepressants genuinely help and can be part of saving a life — the effect is real, replicated, and grows with severity. For mild, everyday misery, the average benefit over placebo is small enough that researchers argue about whether patients can feel it — yet that is exactly where much of the prescribing happens. The drugs were marketed for decades on a "chemical imbalance" story the evidence never supported; withdrawal is real and was long downplayed; and stopping should be planned with a clinician, never done abruptly.
strongBeats placebo for major depression across 500+ trials of 21 drugs — the effect exists; the argument is about size, not existence.
strongHelps more as severity rises: in severe depression the benefit is clinically meaningful; this is where the drugs earn their keep.
strongAlmost certainly prevents relapse in people who have responded — continuation roughly halves relapse rates in trials.
strongWorks for more than depression: solid evidence in anxiety disorders, OCD, and panic disorder.
§ What it doesn't do
Where the marketing outruns the data
strongDoes not do much, on average, for mild depression — the average drug-placebo gap is a few points on a 52-point scale, below common thresholds for what patients notice.
contradictedDoes not fix a "serotonin deficiency," despite a generation of marketing: the 2022 umbrella review found no consistent evidence depression is caused by low serotonin. The drugs can work without the story being true.
strongDoes not work quickly or universally: weeks to onset, and in the largest real-world sequence (STAR*D) roughly a third of patients hadn't remitted after multiple drug steps.
strongThere is no evidence the published record told the whole story: a third of FDA-registered trials went unpublished or were spun — the published literature made the drugs look better than the full data did.
§ The Money Map
Who profits, stage by stage
The golden ageSSRIs were pharma's biggest franchise for two decades. The "chemical imbalance" narrative was a marketing invention that outlived its evidence — it made a prescription feel like insulin for the brain.
The evidence machineIndustry-funded trials, ghost-written papers, and selective publication (documented in the Turner NEJM analysis) inflated apparent efficacy for years.
Today's pipelinePatents expired; generics are cheap. The new money is in telehealth subscription prescribing (questionnaire → refill), and in patented successors (esketamine et al.) priced like the old days.
The other side's moneyAudit the critics too: "get off your meds" content sells courses, supplements, and books. Withdrawal-scare monetization is real, and it kills people who quit abruptly.
This pageNo pharma, no telehealth, no "natural mood" supplements. See "How we make money."
§ The other side
The strongest case against this verdict
The strongest case against our verdict\'s skeptical half: averages hide responders. A drug with a modest average effect can still transform the meaningful minority who respond strongly, and clinicians cannot identify them in advance — so restricting access to "severe only" would deny real help to some mild-moderate patients. Small average effects on symptom scales can also understate functional benefit (getting to work, staying alive). And placebo response in trials is partly the ritual of care — the real-world comparison isn\'t drug vs elaborate placebo ceremony, it\'s drug vs a six-minute GP visit and nothing.
§ Symmetric harms
Harms of using it — and of avoiding it
If you use it
Sexual dysfunction is common, dose-related, and under-disclosed; a subset report it persisting after stopping (PSSD) — acknowledged by EU regulators.
Withdrawal/discontinuation symptoms are real and can be severe and prolonged after long-term use; tapering needs to be slow and planned.
Emotional blunting reported by a substantial minority.
In under-25s: an activation/suicidality signal early in treatment (the FDA black-box warning) — needs close monitoring, not casual telehealth refills.
Elderly: falls, hyponatremia, bleeding interactions.
If you avoid it
Untreated severe depression is lethal: suicide, and a mortality/morbidity burden across every organ system. This is the strongest harm-of-avoiding on this page.
Refusing continuation after response roughly doubles relapse risk.
Quitting abruptly because of something you read — including this page — is the worst of all options: withdrawal + relapse together.
§ The Minority Report
Who gets hurt when it goes right for everyone else
Both tails get hurt: people medicated who barely benefit, and people scared off medication who badly needed it.
Mildly depressed patients on long-term SSRIs: absorbing sexual side effects, blunting, and eventual withdrawal for an average benefit near the noise floor.
Under-25s started via questionnaire telehealth: the black-box group, with the least monitoring.
Undiagnosed bipolar patients: an SSRI alone can precipitate mania — screening is routinely skipped in fast prescribing.
Long-term users never told about withdrawal: some end up in years-long tapers; their suffering was officially minimized for decades.
Severely depressed people talked out of treatment by anti-medication content: the other minority, and the one that dies most.
How to tell if you might be in this group: Ask: how severe is my depression, formally scored? Am I under 25 (monitoring plan)? Any bipolar screening? What is the taper plan for the eventual exit, and what functional change will tell us in 8 weeks whether this is working?
§ The alternatives ledger
Everything else, judged by the same standards
strong
Psychotherapy (CBT and related) Comparable to drugs for mild-moderate depression, with benefits that persist after treatment ends — the durability drugs lack. Access and cost are the honest obstacles.
moderate
Exercise Likely a genuine antidepressant for mild-moderate cases in meta-analyses — unprofitable, so nobody markets it to you.
strong
Lithium, ECT, ketamine — for severe/refractory illness The heavy artillery has strong evidence exactly where SSRIs run out: lithium prevents suicide; ECT works in severe depression; ketamine acts in hours (and is being commercialized accordingly).
moderate
St. John's Wort May work for mild depression (German trials) — and interacts dangerously with many drugs, including birth control and SSRIs themselves. "Natural" is not "inert."
§ The AI Showdown
What other AIs tell you when you ask
We asked each AI the plain question a normal person types: “Do antidepressants actually work?”
ChatGPT
gpt-5.5 · Oct 11, 2026
5/10
A sound, appropriately conditional summary of efficacy that matches the evidence on severity gradient and timeline, but it skips the funding and marketing history, the specific at-risk groups, and concrete questions for a clinician.
●●Matches the evidence. It lands the core conditional verdict — real benefit concentrated in moderate-to-severe depression, smaller placebo gap in mild cases — though it omits the debunked serotonin story and publication-bias record.
●○Names the harms. It lists common side effects (sexual, blunting, weight, early anxiety) and withdrawal, but names no high-risk groups such as under-25s, undiagnosed bipolar, or the elderly, and skips the harms of untreated severe depression beyond a brief self-harm mention.
○○Follows the money. No mention of industry funding, selective publication, chemical-imbalance marketing, telehealth prescribing, or the monetized anti-medication side.
●○Honest about uncertainty. It gives honest conditions and timeframe ranges and avoids false precision, but offers no effect-size context, no cited sources, and doesn't flag that response is unpredictable or that key questions remain open.
●○Useful next step. It advises discussing with a doctor or psychiatrist and warns against abrupt stopping, but supplies no specific questions to ask such as severity scoring, monitoring, or taper planning.
Read its full answer
Yes—antidepressants *do* work for many people, but not for everyone, and the benefit varies.
- **Best evidence:** They are most clearly helpful for **moderate to severe depression**, recurrent depression, and some anxiety disorders.
- **Milder depression:** The difference between antidepressants and placebo is often smaller; therapy, exercise, sleep changes, and addressing stressors may be as effective or better for some.
- **How much they help:** They generally don’t make people feel “artificially happy.” When effective, they reduce symptoms like low mood, anxiety, rumination, sleep/appetite changes, and loss of interest.
- **Timing:** They often take **2–6 weeks** to start helping, and sometimes **6–8+ weeks** for full effect.
- **Trial and error:** If one doesn’t work, another may. People can respond very differently to SSRIs, SNRIs, bupropion, mirtazapine, etc.
- **Side effects:** Common ones include nausea, sleep changes, sexual side effects, emotional blunting, weight changes, or increased anxiety early on. Some are temporary; some aren’t.
- **Stopping:** They shouldn’t usually be stopped abruptly, because withdrawal/discontinuation symptoms can happen.
A fair summary: **antidepressants are real treatments with real evidence, especially for more severe depression, but they’re not magic and they’re not universally effective.** The best outcomes often come from combining medication with psychotherapy, lifestyle changes, and treatment of underlying issues like trauma, substance use, sleep disorders, or medical problems.
If this question is personal, it’s worth discussing with a doctor or psychiatrist—especially if depression is severe, persistent, or involves thoughts of self-harm.
Gemini
gemini-3.8-flash · Oct 11, 2026
4/10
A clear, broadly accurate explainer that lands the severity gradient and the serotonin-myth correction, but it overclaims with unsourced percentages, skips withdrawal and the under-25 warning entirely, ignores the funding and publication-bias story, and leaves the reader without concrete questions for a clinician.
●○Matches the evidence. It correctly cites Cipriani 2018, the severity gradient and the collapse of the serotonin story, but it opens with a flat "yes, antidepressants do work," offers naked 40–60%/30–40% figures without source or conditions, omits publication bias (Turner) and STAR*D non-remission, and asserts neuroplasticity as the mechanism with more confidence than the evidence allows.
●○Names the harms. It lists common side effects (emotional blunting, weight gain, sexual dysfunction) but treats them mainly as reasons people quit, and omits withdrawal/tapering, the under-25 suicidality warning, bipolar activation, and elderly risks.
○○Follows the money. No mention of industry funding, selective publication, chemical-imbalance marketing, or telehealth prescribing incentives — it calls the serotonin story a "myth" without saying who sold it.
●○Honest about uncertainty. It does convey trial-and-error, severity-dependence and placebo response, but presents unsourced percentage ranges as settled and doesn't flag the live debate over whether the average effect is clinically noticeable.
●○Useful next step. It sensibly says medication plus therapy works best and doesn't tell anyone to start or stop anything, but gives no questions to ask a clinician and never explicitly points the reader to a doctor.
Read its full answer
The short answer is **yes, antidepressants do work**, but the longer answer is more nuanced: **they don’t work for everyone, they aren’t a magic cure, and their effectiveness depends heavily on the individual and the severity of their condition.**
Here is what the scientific evidence, clinical trials, and medical consensus tell us about how and why they work.
---
### 1. What the Research Shows
The most comprehensive research to date—including a landmark 2018 meta-analysis published in *The Lancet* that analyzed 522 trials covering over 116,000 people—found that **all 21 common antidepressants studied were more effective than a placebo.**
However, the margin of effectiveness is an important detail:
* **Response rates:** Typically, about **40% to 60%** of people who take an antidepressant see a significant improvement in their symptoms (usually defined as a 50% or greater reduction in symptom severity) within 6 to 8 weeks.
* **The Placebo Effect:** In the same trials, about **30% to 40%** of people who took a dummy pill (placebo) *also* improved.
Antidepressants consistently beat placebos, but the brain’s expectation of healing also plays a massive role in recovery.
---
### 2. Severity Matters
Antidepressants are not equally effective for all types of depression:
* **Severe Depression:** The evidence is strongest here. For people with severe, debilitating depression, antidepressants provide substantial and often life-saving relief compared to a placebo.
* **Mild to Moderate Depression:** The difference between an antidepressant and a placebo is much smaller. For mild depression, clinical guidelines often recommend psychotherapy (like Cognitive Behavioral Therapy) or lifestyle interventions (sleep, exercise, social connection) as the first line of treatment.
---
### 3. Why People Often Think They "Don't Work"
Many people try an antidepressant and feel it failed. There are legitimate medical reasons for this:
* **It’s a trial-and-error process:** Brain chemistry is complex. A medication that works wonders for one person might do nothing—or cause intolerable side effects—for another. It is very common to have to try two or three different medications (or adjust doses) to find one that works.
* **They take time:** Unlike pain relievers or anxiety medications like Xanax, antidepressants do not work immediately. They typically take **2 to 6 weeks** to produce noticeable changes in mood.
* **The "Chemical Imbalance" myth:** Science has moved away from the simplistic idea that depression is just a "shortage of serotonin." Modern neuroscience suggests antidepressants work by promoting **neuroplasticity**—helping the brain grow new connections and adapt over time—which takes time and mental effort.
* **Side effects:** Common side effects (emotional blunting, weight gain, fatigue, sexual dysfunction) lead many people to stop taking them before the therapeutic benefits can take hold.
---
### 4. What Does "Working" Actually Look Like?
Antidepressants are not "happy pills." They do not create artificial euphoria.
When they work properly, they generally:
* Raise the "floor" of your mood so you don't sink as low.
* Reduce overwhelming despair or anxiety enough that you can get out of bed, eat, and function.
* Restore enough energy and mental clarity so you can participate in therapy, exercise, and daily life.
---
### The Bottom Line
Antidepressants are effective medical tools, but they work best as **part of a broader plan**.
The combination of **antidepressant medication plus psychotherapy** has consistently been shown to be more effective than either treatment alone. If someone is struggling, antidepressants can be vital in creating the mental stability needed to make lasting psychological and lifestyle changes.
Claude
claude-opus-5 · Oct 11, 2026
7/10
A well-calibrated, evidence-matched account of the efficacy debate that is weakest on cataloguing side effects and at-risk groups, on the funding/marketing incentives on both sides, and on steering the reader toward a clinician conversation.
●●Matches the evidence. It lands on the same conditional bottom line — real but modest average effect, strongest in severe depression, narrow gap in mild — and correctly separates the serotonin-theory debunk from efficacy, though it omits relapse prevention and STAR*D.
●○Names the harms. It flags withdrawal and slow tapering clearly but omits sexual dysfunction, emotional blunting, the under-25 suicidality warning, bipolar activation, and elderly risks.
●○Follows the money. It cites publication bias via Turner and calls the serotonin story 'oversimplified marketing,' but never names industry funding, telehealth prescribing, or the monetized anti-medication side.
●●Honest about uncertainty. It gives conditions and caveats (effect size on a 52-point scale, NICE threshold, blinding failure, short trials, heterogeneity presented as 'if that's right') without false precision.
●○Useful next step. It offers useful framing about adequate trials and slow tapering but gives no explicit prompt to work with a clinician or concrete questions to ask.
Read its full answer
Short answer: yes, but the honest version is more complicated than either "they're lifesavers" or "they're glorified placebos."
**What the evidence shows**
The largest analysis to date (Cipriani et al., *Lancet* 2018) pooled 522 trials with over 116,000 participants and found all 21 antidepressants studied outperformed placebo. That finding is pretty robust and has survived a lot of scrutiny.
The catch is effect size. The average difference between drug and placebo is around 2 points on the Hamilton Depression Rating Scale, a 52-point instrument. NICE once set 3 points as the threshold for clinical significance. So on average, the drug-minus-placebo benefit is real but modest — roughly a standardized effect of 0.3.
**Why "on average" is doing heavy lifting**
That average probably conceals real heterogeneity. A 2022 BMJ analysis by Stone and colleagues used individual patient data from FDA trials and found the response distribution wasn't uniform — roughly 15% of patients appeared to get a substantial benefit beyond placebo, while most got little. If that's right, the modest average reflects a mix of strong responders and non-responders rather than everyone getting a small nudge. This matches what many clinicians report: some patients describe a dramatic change, others cycle through several drugs with nothing.
**Legitimate criticisms**
- *Publication bias.* Turner et al. (2008) showed 94% of published trials were positive versus only 51% of all FDA-registered trials. Effect sizes in the literature were inflated by about a third.
- *Blinding failure.* Side effects can reveal who's on the active drug, potentially inflating apparent benefit.
- *Short trials.* Most run 6–8 weeks, while people often take these drugs for years, with limited long-term efficacy data.
- *Withdrawal.* This was underplayed for a long time. Discontinuation effects are real and can be severe and prolonged for some people, which complicates both stopping and interpreting relapse-prevention trials.
**A common confusion worth clearing up**
The 2022 Moncrieff review finding little support for the "chemical imbalance"/low-serotonin theory of depression got widespread coverage as evidence that antidepressants don't work. It isn't. Mechanism and efficacy are separate questions — aspirin worked for decades before anyone understood prostaglandins. The serotonin story was always oversimplified marketing, but debunking it doesn't tell you whether the drugs help.
**Practical upshot**
For severe depression, the evidence for benefit is strongest and the case for trying medication is solid. For mild depression, the drug-placebo gap narrows considerably and psychotherapy has comparable evidence with different tradeoffs. Combination treatment generally beats either alone for moderate-to-severe cases.
If a given antidepressant isn't working after an adequate trial at an adequate dose, that's meaningful information — it doesn't mean none will, but it also doesn't mean you should stay on something that isn't helping. And tapering slowly matters more than the older guidance suggested.
Each answer is scored blind against this verdict on five public criteria (2 points each): matches the evidence, names the harms, follows the money, honest about uncertainty, useful next step. The scorer is health.ai's pipeline, which runs on Claude — also a contestant here — so the full answers are shown and a person checks every score before it goes live. Think a score is wrong? Challenge it. AI answers change over time; each card shows the date it was asked. See the full leaderboard →
§ Take it to a human
Questions for your doctor
What is my measured severity, and what does the evidence say for that severity specifically?
Can we try psychotherapy or exercise first — or alongside?
What is the exit plan: when do we reassess, and how would we taper?
Have I been screened for bipolar disorder before starting an SSRI?
Who responds and why — no reliable predictor exists; prescribing remains trial-and-error.
True prevalence of persistent post-treatment sexual dysfunction.
Long-term (decade+) outcomes of continuous use versus episodic treatment.
§ Corrections
Updates & corrections — public, dated, proud
No corrections yet. Found an error? Challenge it — substantiated challenges are corrected within 48 hours and logged here permanently. Submit a challenge →
§ We dissect ourselves too
How WE make money
This pageNo pharmaceutical advertising, no telehealth referral fees, no supplement affiliates. Health.AI earns from optional Pro subscriptions. If that changes, this box changes first.
Demand the next dissection
Any disease, drug, supplement, or practice. The queue decides what we dissect next.
If you're struggling right now: this page is about evidence, not about you — and no verdict is a reason to face a crisis alone. India: iCall 9152987821 · Tele-MANAS 14416. US: call or text 988. Elsewhere: your local emergency number. If you're on medication and want to stop, do it with a clinician — never abruptly.