Sertraline (Zoloft): what the evidence actually shows

Sertraline — sold as Zoloft — is one of the most prescribed antidepressants on earth. That makes it one of the most studied, which is genuinely good news: there is real evidence to look at rather than marketing.

This page walks through what that evidence found, with each claim linked to the study summary behind it. Every linked summary is free to read with no account, and carries a trust score calculated from recorded facts about the paper rather than an editor's opinion.

This is information to bring to your prescriber, not medical advice and not a reason to change anything on your own.

How well does it work?

The reference point is the largest antidepressant comparison ever conducted: a network meta-analysis of 21 antidepressant drugs, published in The Lancet in 2018. Sertraline scored well on both effectiveness and tolerability [1] — the combination is why it turns up so often as a first choice rather than a fallback.

Study summary: Comparative efficacy and acceptability of 21 antidepressant drugs — gold standard, 100/100

The caveat recorded on that resource is worth reading rather than skipping: the analysis was publicly funded (NIHR and the Japan Society for the Promotion of Science), but some co-authors disclosed lecture or consultancy fees from drug manufacturers. That is disclosed, not disqualifying — and it is the kind of thing you should be able to see without digging through a supplement.

A Cochrane review comparing sertraline specifically against other antidepressants adds a practical detail: sertraline causes diarrhoea more often than similar drugs [2]. Its caveat is also worth knowing — many included trials were funded by sertraline's manufacturer, and the review authors ran sensitivity analyses excluding those studies.

Study summary: Sertraline versus other antidepressive agents for depression — gold standard, 100/100

How long before you can tell if it is working?

This is where expectations go wrong most often.

The PANDA trial — a pragmatic, double-blind, placebo-controlled randomised trial in UK primary care — found that sertraline reduced anxiety symptoms before it reduced low mood [3]. The effect on depressive symptoms was slower and less clear-cut in the short window.

Study summary: The clinical effectiveness of sertraline in primary care (PANDA) — gold standard, 91/100

Two useful consequences. First, if you started sertraline for depression and after two weeks feel less anxious but not less flat, that is a recognised pattern rather than a sign it is failing. Second, "give it a few weeks" is not a brush-off — it is what the trial data actually supports. Agreeing a specific review date with your prescriber beats guessing.

The side effect nobody brings up first

Sexual side effects are common across SSRIs and are consistently under-discussed in the appointment.

A meta-analysis of treatment-emergent sexual dysfunction found rates ranging from 25.8% to 80.3%, depending on the antidepressant and, crucially, on how researchers asked [5]. The band is enormous partly because studies that wait for patients to complain spontaneously find far less than studies that use a questionnaire.

Study summary: Treatment-emergent sexual dysfunction related to antidepressants — moderate evidence, 63/100

The caveat on that one is a good example of a paper being honest about itself: the authors state that including open-label studies and pooling across different sexual-function scales "could reduce the significance of our findings."

There is also a regulatory development most people have never heard of. In 2019 the European Medicines Agency's Pharmacovigilance Risk Assessment Committee recommended that labels warn sexual dysfunction can persist after stopping an SSRI or SNRI [6].

Study summary: PRAC recommendations, May 2019 — persistent sexual dysfunction — strong evidence, 70/100

Read that carefully in both directions. A label change is a precautionary act on a safety signal — it is not proof of incidence, causation, or permanence, and the resource says so explicitly. But it is also not nothing, and it is a fair thing to ask about before starting.

For a fuller treatment of this specific question, we have a dedicated page: SSRI sexual side effects: what the research actually found.

What happens when you stop

A randomised trial published in the New England Journal of Medicine compared maintaining antidepressants against discontinuing them in primary care, and found a real relapse risk in the discontinuation group [4].

Study summary: Maintenance or Discontinuation of Antidepressants in Primary Care — gold standard, 87/100

This is not an argument for staying on medication forever. It is an argument for stopping deliberately — with a taper plan and scheduled check-ins agreed in advance — rather than stopping because you felt fine in June.

What the approval was actually based on

Separate from the trials above, there is the regulatory record: what the FDA approval itself rested on, which indication it covers, and what the label says. We publish that per medication, quoted from the record rather than paraphrased.

See the sertraline approval journey. If you want the wider pattern — how long psychiatric drug trials ran before approval, and who paid for them — those are corpus-wide audits: trial durations and research funding.

Questions worth taking to your prescriber

Each of these comes from a specific finding above:

• Sertraline scored well on effect and tolerability in the big comparison — is that part of why you chose it for me?
• How many weeks should we wait before judging whether it is working, given anxiety symptoms may improve before mood?
• If diarrhoea happens, how long does it usually last and when would we switch?
• Where does sertraline sit on the sexual side-effect ranking, and is there an option with a lower rate that would still treat my condition?
• If I ever want to stop, what taper plan and check-ins would you want in place first?

Why this page shows its working

Every study summary linked here is free to read on the web with no account. Each carries a trust score calculated from recorded facts about the paper — study design, funding, journal, sample size, preregistration — recomputed on every read rather than stored, so a rubric correction applies retroactively to the whole library. Where a study has a specific weakness, a written caveat sits next to the score.

The scoring rubric is public. Browse the evidence library and check our work.

If you are in crisis right now, please do not wait: call or text 988 (Suicide & Crisis Lifeline) or text HOME to 741741 (Crisis Text Line). Both are free, confidential and staffed 24/7.