Sertraline: Independent Evidence on Depression, Side Effects & Withdrawal

Table of contents

• Evidence summary
• What sertraline is
• How it works
• What it is prescribed for
• What works and what does not
• Benefits by claim
• Risks and all side effects
• All interactions
• Who should avoid sertraline
• Dosage and how to take it
• Follow the money: who makes it and who funded the evidence
• Related research
• Frequently asked questions
• Sources and funding notes

Evidence summary

The table below ranks the main claims about sertraline by the strength of the human evidence, with funding and conflicts traced for each key source.

Independent evidence and credibility scorecard

Tier 1 is fully independent (a regulator or a publicly funded body with no product stake). Tier 4 is funded or run by the manufacturer. Credibility A–D combines study design, transparency and risk of bias.

What sertraline is

Sertraline is a selective serotonin reuptake inhibitor (SSRI) antidepressant. It is available only on prescription, and its common UK brand name is Lustral (NHS). In the US it is sold as Zoloft. The FDA first approved it on 30 December 1991 (Drugs@FDA, NDA 019839), and the current label lists "Initial U.S. Approval: 1991" (FDA label).

Pfizer (headquartered in New York, USA) originally marketed sertraline as Zoloft. The brand now belongs to Viatris (headquartered in Canonsburg, Pennsylvania, USA), which was formed in November 2020 when Pfizer's off-patent "Upjohn" business was combined with Mylan (SEC Form 8-K, November 2020, SEC EDGAR company record). Viatris is now the licence holder for both US Zoloft and UK Lustral (UK SmPC). Zoloft lost US exclusivity at the end of June 2006, and generic competition began in mid-July 2006 (Pfizer 2006 Financial Report). Today most prescriptions are filled with low-cost generics made by many manufacturers.

It comes as 25 mg, 50 mg and 100 mg tablets. In the US there is also a 20 mg/mL oral concentrate that contains 12% alcohol and must be diluted before use (FDA label). It is not a controlled substance. In a human abuse-liability study it did not produce the euphoria or "drug liking" seen with alprazolam or d-amphetamine (FDA label §9).

How it works

Sertraline blocks the reuptake of serotonin (5-HT) into nerve cells, which increases serotonergic activity in the central nervous system. In vitro it has "only very weak effects" on noradrenaline and dopamine reuptake. It has no significant affinity for adrenergic, cholinergic, GABA, dopaminergic, histaminergic or benzodiazepine receptors, which partly explains why it is less sedating than older tricyclics (FDA label §12). The NHS explains that it is "thought to work" by raising serotonin levels (NHS). The wording is deliberately cautious, because the full reason SSRIs relieve depression and anxiety is not settled.

What the body does with it: peak blood levels occur 4.5–8.4 hours after a dose, and the average elimination half-life is about 26 hours. Steady state is reached after about one week of daily dosing. The main metabolite, N-desmethylsertraline, is much less active and has a half-life of 62–104 hours. Sertraline is 98% bound to plasma proteins. In older people clearance is about 40% lower, so steady state takes 2–3 weeks. In mild liver impairment, exposure was about 3-fold higher (FDA label §12.3). Because the half-life is 24 hours, dose changes should be at least a week apart (UK SmPC).

What it is prescribed for

Licence sources: UK SmPC section 4.1, FDA label section 1. The NHS lists depression, OCD, panic disorder, PTSD and social anxiety disorder (NHS).

What works and what does not

Benefits by claim

Depression: Grade A for efficacy, with honest limits

The 2018 Lancet network meta-analysis led by Oxford's Andrea Cipriani is the most comprehensive independent synthesis available. It pooled 522 double-blind trials (116,477 participants) run between 1979 and 2016, including 86 unpublished studies taken from registries and company websites. Every antidepressant beat placebo. Response odds ratios ranged from 2.13 (95% CrI 1.89–2.41) for amitriptyline down to 1.37 (1.16–1.63) for reboxetine. The authors concluded that escitalopram, mirtazapine, paroxetine, agomelatine "and sertraline had a relatively higher response and lower dropout rate than the other antidepressants". They also wrote that "the summary effect sizes were mostly modest" (Cipriani et al., Lancet 2018). The earlier 2009 analysis of 12 newer antidepressants concluded that sertraline "might be the best choice when starting treatment for moderate to severe major depression in adults because it has the most favourable balance between benefits, acceptability, and acquisition cost" (Cipriani et al., Lancet 2009). A Cochrane review of 59 mostly low-quality head-to-head studies found "a trend in favour of sertraline" but also more diarrhoea than with comparators (Cipriani et al., Cochrane 2010).

How big is the benefit? The per-drug sertraline-versus-placebo odds ratio in the 2018 paper is shown only in a figure, which we could not extract as text, so we do not quote it. The class-level context matters more. An independent Danish meta-analysis of 131 SSRI trials found an average advantage of 1.94 points on the 52-point Hamilton Depression Rating Scale. That falls below the authors' predefined 3-point threshold for clinical significance. The same analysis found that serious adverse events were more common with SSRIs (31 vs 22 per 1,000) and judged that "the clinical significance seems questionable" (Jakobsen et al., BMC Psychiatry 2017). Reasonable experts disagree about where the threshold for clinical significance should sit. In practice, averages hide variation: some people respond well, some not at all, and placebo response in depression trials is substantial.

The PANDA trial: an independent test in real primary care

Most licensing trials recruit people with clearly diagnosed, often moderate-to-severe depression. PANDA was different. It was a UK, NIHR-funded, double-blind trial across 179 GP surgeries that enrolled 655 adults with depressive symptoms of any severity where GPs were genuinely uncertain whether an antidepressant would help. At 6 weeks, mean PHQ-9 scores were 7.98 on sertraline vs 8.76 on placebo, an adjusted proportional difference of 0.95 (95% CI 0.85–1.07; p=0.41). Anxiety (GAD-7) was 21% lower (0.79, 0.70–0.89), mental-health quality of life was higher, and by 12 weeks depression scores were 13% lower (0.87, 0.79–0.97). The authors concluded that sertraline "is unlikely to reduce depressive symptoms within 6 weeks in primary care", but that the anxiety and quality-of-life improvements "are likely to be clinically important" (Lewis et al., Lancet Psychiatry 2019). This trial is valuable because it had no commercial stake and published a null primary result.

Relapse prevention: Grade A

ANTLER (NIHR-funded, 150 UK practices) randomised 478 people who felt well enough to stop citalopram, fluoxetine, sertraline or mirtazapine to either continue or taper onto placebo. Relapse by 52 weeks was 39% with continuation vs 56% with discontinuation (HR 2.06, 95% CI 1.56–2.70). The discontinuation group also had more depression, anxiety and withdrawal symptoms (Lewis et al., NEJM 2021). The UK SmPC recommends at least 6 months of treatment after depression resolves (SmPC).

OCD, PTSD, panic, social anxiety and PMDD: Grade B

For OCD, SSRIs as a class reduced YBOCS scores by 3.21 points versus placebo and increased response (RR 1.84), with no statistical difference between individual SSRIs (Soomro et al., Cochrane 2008). For PTSD, SSRIs improved symptoms in 58% vs 35% on placebo (moderate certainty), and the reviewers call them "first-line agents for the pharmacotherapy of PTSD" (Williams et al., Cochrane). The Pfizer-funded pivotal sertraline PTSD trial reported 53% responders vs 32% on placebo (Brady et al., JAMA 2000). Yet the regulator's label records that two further placebo-controlled PTSD trials were not statistically significant, and that post-hoc analyses found "essentially no effect in the relatively smaller number of men" (FDA label §14.4). For panic disorder, seven people need antidepressant treatment for one extra person to benefit (Bighelli et al., Cochrane 2018). For PMS/PMDD, the SSRI effect was moderate (SMD −0.65) when pooling end scores and small (SMD −0.36) when pooling change scores (Marjoribanks et al., Cochrane 2013).

Depression after a heart attack: safety data

SADHART randomised 369 people with major depression after a heart attack or unstable angina. Sertraline did not change heart pumping function (LVEF), ventricular ectopy or QTc compared with placebo, and severe cardiovascular events occurred in 14.5% vs 22.4% (not statistically significant). On depression, the clinician global rating favoured sertraline, but the primary depression scale (HAM-D) did not reach significance in the whole sample (P=.14) (Glassman et al., JAMA 2002). The trial was funded mainly by Pfizer, with a Pfizer employee involved throughout (funding summary).

Risks and all side effects

FDA boxed warning: suicidal thoughts and behaviours

The US label carries a boxed warning: "Antidepressants increased the risk of suicidal thoughts and behavior in pediatric and young adult patients in short-term studies." All treated patients should be closely monitored for clinical worsening and new suicidal thoughts. The pooled FDA data (about 77,000 adults and more than 4,400 children) show 14 additional cases per 1,000 treated under 18 and 5 additional cases per 1,000 aged 18–24. They show 1 fewer case per 1,000 aged 25–64 and 6 fewer per 1,000 aged 65 and over (FDA label §5.1). The FDA's own adult analysis found risk "strongly age dependent" (Stone et al., BMJ 2009). NICE recommends a review 1 week after starting for people aged 18–25 or at particular suicide risk (NICE NG222). In December 2025 the MHRA announced that patient leaflet wording on suicidal behaviour for 28 antidepressants would be strengthened, with a patient card to be introduced (MHRA, 1 December 2025).

Common side effects

The NHS lists headaches, nausea and vomiting, dizziness or drowsiness, dry mouth, diarrhoea, problems sleeping, sexual problems and weight gain as common. Most ease "after a couple of weeks", although some last longer (NHS). In pooled placebo-controlled trials (3,066 people on sertraline, 2,293 on placebo, 8–12 weeks), 12% stopped sertraline because of an adverse reaction, compared with 4% on placebo (FDA label §6.1).

Source: FDA label, Tables 3 and 4. These are manufacturer trial data reviewed by the FDA.

Serious risks and regulator warnings

Withdrawal (discontinuation) symptoms

Sertraline is not addictive in the sense of craving or euphoria, but the body adapts to it, and stopping, especially abruptly, can cause withdrawal symptoms. The UK SmPC reports that in clinical trials, withdrawal reactions occurred in 23% of people discontinuing sertraline vs 12% of those continuing. The most common symptoms are dizziness, sensory disturbances (including "electric shock" sensations), sleep disturbance and intense dreams, agitation or anxiety, nausea, tremor and headache. They usually resolve within 2 weeks, "though in some individuals they may be prolonged (2-3 months or more)" (UK SmPC).

How common is it across antidepressants? The evidence is contested:

• A 2024 meta-analysis of 79 studies (21,002 patients) found discontinuation symptoms in 31% after stopping an antidepressant vs 17% after stopping placebo. It estimated about 15% as attributable to the drug, and severe symptoms in 2.8% vs 0.6% (Henssler et al., Lancet Psychiatry 2024; no funding).
• A 2019 review reported withdrawal incidence ranging from 27% to 86% (weighted average 56%). It found that a significant proportion of people experience symptoms for more than two weeks, and urged guideline updates (Davies & Read, Addict Behav 2019).
• An influential 2019 analysis of serotonin-transporter occupancy argued that SSRIs should be tapered "hyperbolically and slowly to doses much lower than those of therapeutic minimums" (Horowitz & Taylor, Lancet Psychiatry 2019).

NICE revised its guidance in 2022. It now tells patients that withdrawal symptoms usually go away within 1 to 2 weeks, but can sometimes last several weeks and "occasionally several months", and can be severe after sudden stopping (NICE NG222 1.4.14).

Pregnancy and newborn effects

According to the UK SmPC, "a substantial amount of data did not reveal evidence of induction of congenital malformations by sertraline". Newborns exposed late in pregnancy may develop withdrawal-type or serotonergic symptoms, usually within 24 hours of birth. SSRI use late in pregnancy has been linked to persistent pulmonary hypertension of the newborn (PPHN), at about 5 cases per 1,000 pregnancies against a background of 1 to 2 per 1,000 (UK SmPC 4.6). The UK Bumps service puts the PPHN figure at around one in every 300 babies whose mother takes an SSRI. It also stresses that untreated depression in pregnancy carries its own risks (Bumps).

All interactions

Interaction sources: FDA label sections 4, 5 and 7, UK SmPC sections 4.3–4.5, NHS. The label lists no clinically important interaction with cimetidine, diazepam, lithium pharmacokinetics, atenolol, tolbutamide, digoxin or CYP3A4 substrates.

Who should avoid sertraline

Must not take it (contraindications): people taking, or within 14 days of stopping, an MAOI (including linezolid or IV methylene blue); people taking pimozide; anyone with known hypersensitivity to sertraline, such as previous anaphylaxis or angioedema; and people taking disulfiram, who must not use the alcohol-containing US oral solution (FDA label §4, UK SmPC 4.3).

Needs a careful prescriber decision: the NHS says it may not be suitable for people with diabetes, epilepsy, heart disease, glaucoma, a bleeding disorder (particularly in the digestive system) or a history of mania (NHS). Other groups needing care are people with QT risk factors (FDA §5.10) and people taking blood thinners or NSAIDs.

• Liver disease: the US label recommends half the usual dose in mild impairment and does not recommend sertraline in moderate or severe impairment. The UK SmPC says it should not be used in severe hepatic impairment (FDA §8.6, SmPC 4.2).
• Kidney disease: no dose adjustment is needed (FDA §8.7).
• Older adults: clearance is about 40% lower, so dosing should start "at the low end of the dosing range". Hyponatraemia is more likely, and SSRIs are associated with more fractures in people over 50 (FDA §8.5, SmPC). The American Geriatrics Society Beers Criteria also address SSRIs in older adults. We could not retrieve the full 2023 text for this review, so we do not quote it.
• Children and teenagers: licensed only for OCD (ages 6–17) in both the UK and US. The UK SmPC says sertraline should not otherwise be used under 18. Growth and weight should be monitored (SmPC 4.4, FDA §8.4).
• Pregnancy: the NHS says sertraline "can be used during pregnancy if needed", at the lowest effective dose, usually with a hospital birth so mother and baby can be monitored (NHS). The UK SmPC is more restrictive in wording ("not recommended… unless" the benefit outweighs the risk). Bumps advises that women should not stop sertraline without first speaking to their midwife, GP or specialist (Bumps).
• Breastfeeding: in a pooled analysis of 53 mother–infant pairs, exclusively breastfed infants had on average 2% (range 0–15%) of the mother's serum level, and no adverse reactions were seen (FDA §8.2). The NHS says it "can often be used while breastfeeding" (NHS).

Dosage and how to take it

Your prescriber sets and adjusts your dose. The table shows only the official licensed adult ranges for reference. It is not a dosing guide.

Dose increases, when needed, are made in steps (the UK SmPC says 50 mg; the US label says 25–50 mg) at intervals of at least one week (UK SmPC 4.2, FDA §2). The US label halves the dose in mild liver impairment.

How it is taken: once a day at the same time, morning or evening, with or without food. If you miss a dose, skip it and take the next dose as usual; do not double up. Taking more than prescribed can be dangerous, so get help from NHS 111 (NHS).

How long before it works: the UK SmPC says an effect "may be seen within 7 days" but "longer periods are usually necessary", especially in OCD (SmPC). NICE says benefits "should be felt within 4 weeks". It advises a review usually within 2 weeks of starting, and a discussion of next steps if there has been no response at all after 4 weeks at a therapeutic dose (NICE NG222). Treatment for depression is typically continued for at least 6 months after symptoms resolve (SmPC).

How to stop: only with your prescriber. The NHS says your doctor will reduce the dose gradually "over several weeks or months" (NHS). NICE advises reducing "in a step-wise fashion, at each step prescribing a proportion of the previous dose (for example, 50% of previous dose)". It suggests smaller reductions (for example 25%) as the dose gets lower, and liquid preparations if tablets cannot achieve small enough doses. The pace should be agreed with the patient, and any withdrawal symptoms should be resolved or tolerable before the next step (NICE NG222 1.4.16). If intolerable symptoms occur, the SmPC allows going back to the previous dose and then reducing more gradually (SmPC 4.2).

Follow the money: who makes it and who funded the evidence

• Originator and current owner: Pfizer Inc (New York, USA) originally held the Zoloft licence. The brand passed to Viatris Inc (Canonsburg, Pennsylvania, USA) when Pfizer's Upjohn off-patent business combined with Mylan on 16 November 2020. At that point Pfizer shareholders owned about 57% of Viatris (SEC 8-K, SEC EDGAR). Viatris Products Limited holds the UK Lustral licence (UK SmPC).
• Revenue: Pfizer reported Zoloft revenues of $3,361 million in 2004 and $3,256 million in 2005. After US exclusivity ended in June 2006, revenue fell to $2,110 million in 2006, a 35% drop (Pfizer 2006 Financial Report).
• Generics: sertraline is now off-patent and made by many companies. FDA records list abbreviated (generic) applications from firms including Aurobindo Pharma, Lupin, Zydus Lifesciences, Accord Healthcare and Granules (openFDA Drugs@FDA data). Low generic prices are part of why NICE calls sertraline "the most cost-effective drug" for GAD (NICE CG113).
• Who funded the licensing trials: the efficacy studies in the FDA label (MDD, OCD, panic, PTSD, social anxiety, PMDD) are the sponsor's regulatory trials (FDA label §14). For example, the pivotal PTSD trial states "This study was funded by Pfizer Inc". Two of its authors were Pfizer employees who held Pfizer stock, and the lead author sat on a Pfizer advisory board (Brady et al., JAMA 2000, full text). SADHART was likewise Pfizer-funded (funding summary). Manufacturer funding does not make these results false, and the FDA label itself discloses the negative PTSD trials.
• Who funded the independent evidence: the UK's National Institute for Health Research funded PANDA and ANTLER and co-funded the 2018 Cipriani network meta-analysis, with no funder role in analysis or reporting (PANDA, ANTLER, Cipriani 2018). The FDA's suicidality meta-analysis was done by agency staff with no specific grant (Stone 2009). The Henssler withdrawal meta-analysis and the Slee GAD network meta-analysis report no funding (Henssler 2024, Slee 2019).
• Residual conflicts even in independent work: in Cipriani 2018, one co-author declared lecture fees from companies including Pfizer, and another declared consulting or lecture honoraria from several manufacturers. In PANDA, one author reported a Pfizer Independent Grants for Learning and Change award outside the submitted work (PANDA declarations). These are disclosed ties, not evidence of bias.
• Industry share of the trial base: Cipriani's team found that industry funding "was not associated with substantial differences" in response or dropout, but cautioned that "non-industry funded trials were few and many trials did not report or disclose any funding" (Cipriani 2018).
• Documented integrity concern: a 2003 British Journal of Psychiatry study compared sertraline articles "coordinated by a medical writing agency" with other sertraline articles. The agency-linked articles had more authors (6.6 vs 2.95 per article) and far higher citation rates (20.2 vs 3.7). The authors concluded this style of authorship "raises concerns for the scientific base of therapeutics" (Healy & Cattell, Br J Psychiatry 2003). The study describes a pattern of publication influence. It did not find that any particular sertraline result was false.

Related research

• Stress, anxiety and depression: prevention and management guide
• Supplements for stress, anxiety and depression: the evidence
• St John's wort, which must not be combined with sertraline
• Saffron for depression
• Ashwagandha: evidence and safety
• L-theanine and magnesium
• Sleep guide, sleep supplements and melatonin
• Sibling medicine reviews: fluoxetine, escitalopram, mirtazapine, duloxetine

Frequently asked questions

Sources and funding notes

• NHS, Sertraline (page last reviewed 8 September 2025): UK public health service; no product stake.
• FDA-approved Zoloft prescribing information (DailyMed, Viatris Specialty LLC): US regulator-approved label written by the licence holder; includes the boxed warning.
• Drugs@FDA, NDA 019839 (Zoloft): US regulator; original approval 30 December 1991.
• Lustral 50 mg SmPC (electronic medicines compendium): UK MHRA-approved text; licence holder Viatris Products Limited; revised 12/2025.
• NICE NG222, Depression in adults (2022): UK national guideline body.
• NICE CG113, Generalised anxiety disorder and panic disorder: UK national guideline body.
• NICE NG116, Post-traumatic stress disorder (2018): UK national guideline body.
• NICE CG31, OCD and body dysmorphic disorder: UK national guideline body.
• NICE CG159, Social anxiety disorder: UK national guideline body.
• MHRA Drug Safety Update, SSRI/SNRI and postpartum haemorrhage (7 January 2021): UK regulator (executive agency of the Department of Health and Social Care).
• MHRA, antidepressant safety information review (1 December 2025): UK regulator; the Commission on Human Medicines is an independent statutory body.
• MHRA, SSRIs and SNRIs: use and safety (2014, historical record): UK regulator; background on suicidality reviews.
• Bumps, Sertraline in pregnancy (June 2022, v3.1): UK public pregnancy-medicines information service; no product stake.
• Cipriani et al., Lancet 2018 (21-antidepressant network meta-analysis): funded by NIHR Oxford Health BRC and the Japan Society for the Promotion of Science; some co-authors declared pharma fees; UK/Japan/international.
• Cipriani et al., Lancet 2009 (12 new-generation antidepressants): academic multinational team; funding not stated in the abstract.
• Cipriani et al., Cochrane 2010, Sertraline versus other antidepressive agents: Cochrane review; 59 mostly low-quality studies.
• Lewis et al., Lancet Psychiatry 2019 (PANDA trial): NIHR Programme Grant RP-PG-0610-10048; UK; independent; two authors declared Pfizer-linked grants outside this work.
• Lewis et al., NEJM 2021 (ANTLER trial): NIHR-funded; UK; independent.
• Jakobsen et al., BMC Psychiatry 2017 (SSRIs vs placebo): Copenhagen Trial Unit staff support only; Denmark; no competing interests.
• Slee et al., Lancet 2019 (GAD network meta-analysis): "No funding was received"; UK academic authors.
• Soomro et al., Cochrane 2008 (SSRIs for OCD): no conflicts stated.
• Williams et al., Cochrane (Pharmacotherapy for PTSD; search to November 2020): academic authors; lead author funded by the South African Medical Research Council; one author declared industry grants or honoraria.
• Bighelli et al., Cochrane 2018 (antidepressants for panic disorder): academic; one author declared pharma lecture fees; the review notes industry funding of some trials.
• Williams et al., Cochrane 2017 (pharmacotherapy for social anxiety disorder): academic; one author declared pharma grants or honoraria.
• Marjoribanks et al., Cochrane 2013 (SSRIs for premenstrual syndrome): one author declared long-term industry funding.
• Brady et al., JAMA 2000 (sertraline for PTSD): "funded by Pfizer Inc"; two Pfizer-employed authors with Pfizer stock (full-text disclosure); USA; manufacturer-funded.
• Glassman et al., JAMA 2002 (SADHART): funded by Pfizer and foundations, with a Pfizer employee involved (Wiki Journal Club summary); multinational; manufacturer-funded.
• Stone et al., BMJ 2009 (FDA suicidality analysis): FDA staff; no specific grant; no competing interests; USA.
• Henssler et al., Lancet Psychiatry 2024 (incidence of discontinuation symptoms): no funding; German university authors.
• Davies & Read, Addictive Behaviors 2019 (withdrawal review): UK academic authors; affiliations listed include the All-Party Parliamentary Group for Prescribed Drug Dependence and the International Institute for Psychiatric Drug Withdrawal; funding not stated in the abstract.
• Horowitz & Taylor, Lancet Psychiatry 2019 (tapering SSRIs): academic authors (Australia/UK); funding not stated in the abstract.
• Healy & Cattell, Br J Psychiatry 2003 (authorship and industry): UK academic authors; analysis of agency-coordinated sertraline publications.
• Pfizer 2006 Financial Report (SEC filing): company financial disclosure; Zoloft revenues and loss of exclusivity.
• Mylan/Viatris Form 8-K, November 2020 (SEC filing): company disclosure of the Upjohn–Mylan combination.
• SEC EDGAR company records for Viatris and Pfizer: official registered business addresses.

Last reviewed: 26 September 2026. Educational information from Pure City Research, not medical advice. Medicine decisions belong with a qualified prescriber.