What is depression?

Depression is one of the most common mental health conditions worldwide and a leading cause of disability. It is more than low mood: depression is a persistent state that affects how people think, feel, function, and relate to others. Symptoms include low or empty mood, loss of interest or pleasure, fatigue, difficulties with concentration and sleep, feelings of worthlessness, and, in severe cases, thoughts of suicide or self-harm. Depression varies enormously in severity and presentation, from mild and transient episodes to severe and recurring illness that can be disabling.

Depression affects people of all ages, backgrounds, and circumstances. It is often accompanied by anxiety, physical health problems, and social difficulties, and these interactions make both assessment and treatment more complex. Despite high prevalence and significant impact, many people with depression do not receive effective treatment: access barriers, stigma, and gaps in service provision all play a role.

What the research tells us

Our research summaries are based on blogs published since 2018, each written by an expert who has read and critically appraised the original study.

Antidepressants: efficacy, mechanism, and side effects

The efficacy of antidepressants for adult depression is well established. A landmark 2018 network meta-analysis of 522 trials and 116,000 participants confirmed that all 21 antidepressants tested outperformed placebo. However, efficacy alone does not determine which drug is right for which patient. A 2025 systematic review ranking antidepressants by their physical health side effects found meaningful differences between drugs in terms of cardiovascular, metabolic, and other risks, reinforcing that drug choice matters. Concerns about weight gain are also borne out by evidence: a 2019 cohort study found long-term antidepressant use was associated with significant weight gain at the population level.

The debate around the serotonin hypothesis attracted renewed attention following a 2022 review that questioned whether low serotonin underpins depression. A 2023 NES response to that review argued the evidence is more nuanced and that the mechanism debate does not undermine the case for prescribing antidepressants. Separately, a 2023 review on antidepressants, cognition, and emotional blunting found that reduced emotional responsiveness is a real and under-recognised effect, likely reflecting a neurobiological rather than purely psychological process.

A patient-centred question concerns what counts as a meaningful benefit. A 2024 survey study explored the Smallest Worthwhile Difference for antidepressant treatment, finding that patients and clinicians differ in their thresholds, with implications for how trial results are interpreted in practice.

Stopping antidepressants safely

Discontinuation remains one of the most contested and clinically challenging areas of depression care. A 2022 systematic review found that international guidelines on antidepressant discontinuation are unclear and often unhelpful, leaving clinicians without reliable guidance on who should stop, when, and how. A 2024 meta-analysis of discontinuation symptoms found that around 8 to 14 percent of patients experience them, with roughly 2 percent experiencing severe symptoms, a more modest estimate than some published accounts but still clinically significant.

The ANTLER trial provided important primary care data: a 2021 blog on the trial’s findings reported that discontinuing antidepressants substantially increased relapse risk in patients who had been well on medication. This challenges the assumption that stopping is always the appropriate goal. For those who do wish to stop, patient and lived experience perspectives on tapering have been influential, and a 2025 network meta-analysis compared deprescribing strategies, finding that gradual hyperbolic tapering is better tolerated than abrupt discontinuation. A 2024 survey of people with lived experience of antidepressant withdrawal called for better information, slower tapering, and more clinical support during the process.

Psychological therapies

CBT remains the most thoroughly evaluated psychological treatment for depression, but a 2023 systematic review asked whether it is truly the best we have, finding it effective but questioning whether effect sizes justify its privileged position over other evidence-based therapies. A 2022 trial comparing stratified and stepped care approaches found no significant difference in outcomes, raising questions about how treatment should be sequenced in practice.

Preventing relapse is a key clinical priority. A 2024 meta-analysis found psychological interventions are effective in reducing relapse and recurrence, with mindfulness-based cognitive therapy and CBT showing the strongest evidence. A 2025 economic evaluation found behavioural activation to be cost-effective for depression in older adults, an important finding given the high prevalence of depression in later life and the limited reach of standard NHS Talking Therapies provision for this group.

Collaborative care models are gaining interest as a way to integrate physical and mental health support. A 2025 meta-analysis identified the key components of collaborative care for depression, including case management, regular follow-up, and active links between primary and specialist care.

Novel and neuromodulation treatments for treatment-resistant depression

Between 20 and 30 percent of people with depression do not respond adequately to first-line treatments. Evidence for novel options has grown considerably since 2018. A 2023 blog on psilocybin for treatment-resistant depression called it an “island of hope in an ocean of uncertainty,” with trial data showing meaningful response rates but significant questions remaining about durability, safety, and how to deliver it at scale. Ketamine has a more established evidence base: the ‘Ketamine and me’ qualitative project documented patients’ lived experience of ketamine treatment, and a 2023 review on ketamine-assisted therapy summarised the rapidly evolving evidence. Esketamine (intranasal) received regulatory approval during this period, with the first clinical study reviewed on NES in 2019.

Neuromodulation is also an active area. Deep brain stimulation may improve long-term quality of life for treatment-resistant depression, according to a 2022 review, though it remains a specialist intervention. A 2025 review of home-based transcranial direct current stimulation (tDCS) asked whether the technology is ready for widespread use, concluding that the evidence is promising but not yet sufficient to recommend routine implementation. For those who have not responded to antidepressants, results from the 2025 LQD study suggest quetiapine may be marginally preferable to lithium for augmentation, though both remain viable options.

Exercise as an evidence-based treatment

The evidence for exercise in depression has strengthened considerably. A major 2023 systematic review concluded that exercise is a genuine evidence-based treatment option for depression, not merely a lifestyle recommendation. A 2025 network meta-analysis comparing exercise modalities found vigorous aerobic exercise, yoga, and combined aerobic plus resistance training to be most effective, with effect sizes comparable to psychotherapy. A 2024 review confirmed that even small doses of physical activity reduce depression risk, with dose-response benefits extending well below current activity guidelines, which has implications for how clinicians frame exercise recommendations to patients who feel exercise is unachievable.

Digital and technology-based interventions

Digital CBT and app-based interventions have moved from experimental to increasingly mainstream. A 2026 real-world cohort study found that internet-delivered CBT achieved similar outcomes to face-to-face therapy for depression, an important finding for policy and commissioning. A 2024 meta-analysis of apps for depression and anxiety found a positive overall effect, though effect sizes were modest and retention remains a challenge. A 2025 study found that generative AI chatbots can reduce clinically significant mental health symptoms, raising both opportunities and questions about safety, personalisation, and the role of human oversight. A 2023 review as part of the DepressionSolvingTheToll series argued that digital tools offer the best prospect for reaching the majority of people with depression who currently receive no treatment, particularly in low- and middle-income countries.

Depression in young people

Youth depression is not simply a younger version of adult depression. A 2023 systematic review of adolescents’ lived experience of depression found that young people emphasise social, relational, and identity dimensions that are often absent from adult frameworks and adult-derived outcome measures. A 2025 review rethinking the evidence base for young people’s depression treatments argued that much of the trial evidence applies poorly to real-world clinical populations, and called for more nuanced interpretation of effect sizes. A 2021 Cochrane review confirmed that antidepressants should not be the first port of call for youth depression, with psychological therapies preferred, though access to these therapies remains patchy. A 2025 analysis found that young adults benefit less than older adults from NHS psychological therapies, pointing to a need for age-appropriate service design in the 18 to 25 age group.

Perinatal depression

Depression during pregnancy and the postnatal period affects a significant proportion of women, and is associated with adverse outcomes for both parent and child. A 2024 systematic review of perinatal screening practices found that current approaches frequently fall short of guideline recommendations, with inconsistent timing and under-identification of women from minority ethnic groups. A 2023 trial of brief interpersonal therapy for maternal depression during pregnancy found meaningful reductions in depressive symptoms, supporting the use of brief, focused psychological interventions in the antenatal period. A 2025 trial of culturally adapted CBT for British South Asian women with postnatal depression found recovery rates comparable to those seen in the general population when therapy is appropriately adapted. Perinatal depression in fathers and non-gestational parents is also increasingly recognised: a 2025 review on paternal postnatal depression highlighted the evidence base and called for services to proactively identify and support fathers.

Key practice points

• Choice of antidepressant matters beyond efficacy. All antidepressants outperform placebo, but they differ meaningfully in tolerability, physical health side effects, and weight gain risk. Recent evidence should inform shared decision-making about which drug to start.
• Discontinuation needs careful planning. International guidelines remain inconsistent, and discontinuation symptoms affect a meaningful minority of patients. A slow, hyperbolic taper with clinical support is now recommended over abrupt stopping.
• CBT is evidence-based but not the only option. Multiple psychological therapies work for depression. Mindfulness-based cognitive therapy and CBT have the strongest evidence for relapse prevention, and behavioural activation is a cost-effective alternative, particularly for older adults.
• Exercise is a legitimate treatment, not just a lifestyle add-on. Vigorous aerobic exercise, yoga, and combined training all show clinically meaningful effects. Even small amounts of activity reduce depression risk.
• For treatment-resistant depression, a growing toolkit exists. Esketamine, psilocybin (in trials), ketamine, quetiapine augmentation, and brain stimulation technologies all have evidence behind them. A 2024 review summarises the current state of play.
• Digital CBT can be as effective as face-to-face therapy. Real-world evidence now supports internet-delivered CBT as a genuine alternative, not just a stop-gap. Retention remains a challenge across digital platforms.
• Young people need age-appropriate approaches. Adolescent depression is phenomenologically distinct from adult depression. Psychological therapies are preferred over antidepressants as first-line treatment, and trial evidence should be interpreted cautiously when applied to real-world clinical populations.
• Perinatal depression is underdetected and undertreated. Screening is inconsistent, particularly for women from minority ethnic communities. Brief psychological interventions and culturally adapted approaches have good evidence and should be available to all.

Featured blogs

• Antidepressants can help adults with major depression. Andrea Cipriani, Feb 2018. The landmark network meta-analysis of 522 trials and 116,477 participants: the definitive evidence that antidepressants work for major depression in adults.
• Is cognitive behavioural therapy the best we’ve got for depression?. Holly Victoria Pote, Jul 2023. A rigorous systematic review of CBT for depression, interrogating whether it truly deserves its privileged position and what the alternatives offer.
• Exercise for depression: an evidence-based treatment option. Brendon Stubbs, Nov 2023. Comprehensive review placing exercise on equal footing with other evidence-based treatments, with implications for how clinicians discuss physical activity with patients.
• Psilocybin for treatment-resistant depression: an island of hope in an ocean of uncertainty?. Iain Perdue, Oct 2023. Balanced appraisal of trial data on psilocybin, setting out what the evidence supports and what questions remain before clinical implementation.
• Maintenance or discontinuation of antidepressants: findings from the ANTLER trial. Raphael Rifkin-Zybutz, Nov 2021. Primary care RCT data showing that discontinuing antidepressants substantially increases relapse risk. A must-read for anyone advising patients on stopping medication.
• Apples and oranges? Rethinking the evidence behind young people’s depression treatments. Jack Andrews, Mar 2025. A thoughtful challenge to how trial evidence is applied to young clinical populations, raising questions about generalisability that matter for everyday practice.

Explore related topics

Related conditions

Anxiety,
Bipolar disorder,
PTSD,
Self-harm,
Suicide,
Sleep disorders

Related treatments

Antidepressants,
CBT,
Exercise,
Psychotherapy,
Mindfulness,
Neurostimulation,
Digital health

Related populations and settings

Child and adolescent,
Perinatal mental health,
Poverty,
LGBTQ+,
Primary care