View commentaries on this research

This is a plain English summary of an original research article

This review found no difference in effectiveness or drop-out rates between antidepressants and cognitive behavioural therapy for adults recently diagnosed with major depressive disorder. Both treatments should be offered, as recommended by NICE, either alone or possibly in combination, and the final decision will rely heavily on the patient’s preference.

The challenge for talking therapies in the NHS has long been a lack of capacity. However, the Improving Access to Psychological Therapies programme has in the last few years provided thousands of trained therapists who can be accessed through GPs and in some cases directly.

The trials included in this review were prone to bias and often excluded difficult to treat populations, such as the elderly and those with medical comorbidities and at risk of suicide. This review, while a good summary of the current evidence, is therefore unlikely to be the final word on the matter. NIHR has funded some relevant research, summarised in a recent highlight on talking therapies for depression.

Why was this study needed?

There were about 1.44 million consultations for depression in England in 2010, costing the NHS more than £520 million. Depression imposes a large burden on individuals, families and society, reducing quality of life and lowering productivity.

Antidepressants are commonly prescribed for depression. However, it has been reported that one in five patients don't pick up their prescriptions, and of those that do many fail to complete the full course. More than 60% of patients have adverse effects during treatment with a second generation antidepressant, and although most adverse effects are relatively minor (for example, constipation, diarrhoea and dizziness) they may contribute to the decision to stop taking them. Many patients prefer talking therapies, such as cognitive behavioural therapy – quoting concerns about side effects of drugs and the fear that they might become dependent on antidepressants.

To help inform the decision of drug versus talking therapy, this review set out to examine the benefits and harms of the two approaches to treatment.

What did this study do?

This was a systematic review and meta-analysis of 11 randomised controlled trials comparing a second generation antidepressant with cognitive behavioural therapy for the initial treatment of major depressive disorder in adults. Ten trials compared antidepressants with cognitive behavioural therapy alone, while three compared antidepressants with cognitive behavioural therapy plus antidepressants.

The antidepressants used were all second generation; fluoxetine, fluvoxamine, paroxetine, sertraline, venlafaxine, citalopram and escitalopram. Two trials were in primary care, the rest in outpatient settings. Four trials were in the United States, one was in England, and the remaining in Canada, Germany, Iran and Romania.

This was a high quality review. However, all but one of the included trials had medium to high risk of bias, often because of high drop-out rates. Confidence intervals (CI) for some outcomes were wide. Results therefore need to be interpreted with caution.

What did it find?

  • There was no statistically significant difference in effectiveness between antidepressants and cognitive behavioural therapy for either treatment response (risk ratio [RR] 0.91, 95% CI 0.77 to 1.07, from a meta-analysis of five trials) or remission (RR 0.98, 95% CI 0.73 to 1.32, from a meta-analysis of three trials), as measured by changes in the Hamilton Rating Scale for Depression.
  • Adverse events were poorly reported, so discontinuation (drop-out) rates were compared as proxy measures for adverse events. There was no difference in discontinue rates between antidepressants and cognitive behavioural therapy (RR 0.90, 95% CI 0.49 to 1.65).
  • The three trials that compared antidepressant alone with a combination of antidepressant and cognitive behavioural therapy reported no statistically significant differences between groups in rates of remission or response.

What does current guidance say on this issue?

NICE guidance published in 2009 recommends that people with moderate or severe depression should be provided with antidepressant medication, or a high-intensity psychological intervention, or combination therapy. The choice of intervention should be influenced by the: duration of the episode of depression and the trajectory of symptoms; previous course of depression and response to treatment; likelihood of adherence to treatment and any potential adverse effects; person's treatment preference and priorities.

What are the implications?

This review found no significant differences in effectiveness between drug and cognitive behavioural therapies for major depressive disorder – either alone or in combination. However there was no clear comparison of the adverse effects of each treatment due to the quality of the studies. Given that patients may have personal preferences for one treatment over the other, both treatments should be made accessible, either alone or in combination, to patients with major depressive disorder. This is in line with current NICE guidance.

Note that although this review appeared to find no benefit of combining therapies, more recent research in related but distinct area of treatment-resistant depression, featured in the NIHR highlight, suggests benefits of combining talking therapies and medication.

One of the challenges for talking therapies in the NHS is restricted capacity. However the situation is changing for the better. The Improving Access to Psychological Therapies programme has provided GPs with access to thousands of trained therapists, and around half of surgeries in England now provide counselling services and support. GPs should ensure they offer talking therapies when available – a recent survey found that only 42% of people who visited their GP with depression were offered counselling, even though 82% of them would have been willing to try it.

The review was of adults with major depressive disorder. The results are therefore not applicable to children or to people with other conditions such as “sub-threshold” depression, dysthymia or perinatal depression. Also, the included trials consisted mostly of patients with moderate to severe major depressive disorder, so results may not be applicable to people with milder depression. Finally, most trials excluded patients with medical comorbidities or suicidal thoughts, and only a few trials included elderly patients – most trials reported a mean age of between 35 and 45 years. Future trials should try to recruit people who better reflect the general population of people with major depressive disorder.

These results come from a larger comparative effectiveness review of benefits and harms of second generation antidepressants, psychotherapies (including therapies other than cognitive behavioural therapy), complementary and alternative medicine treatments, and exercise interventions for major depressive disorder funded by the US Agency for Healthcare Research and Quality.

Citation and Funding

Amick HR, Gartlehner G, Gaynes BN, et al. Comparative benefits and harms of second generation antidepressants and cognitive behavioral therapies in initial treatment of major depressive disorder: systematic review and meta-analysis. BMJ. 2015;351:h6019.

This project was funded under contract from the Agency for Healthcare Research and Quality by the RTI-UNC Evidence-based Practice Center.


NHS Choices. Can I get free therapy or counselling? London: NHS Choices; 2014.

NICE. Depression in adults: recognition and management. CG90. London: National Institute for Health and Care Excellence; 2009.

NICE. Judging whether public health interventions offer value for money. LGB10. London: National Institute for Health and Care Excellence; 2013.

Produced by the University of Southampton and Bazian on behalf of NIHR through the NIHR Dissemination Centre

  • Share via:
  • Print article


The Hamilton Rating Scale for Depression is a multiple-item questionnaire used to provide a measure of depression in adults. It was originally published in 1960 and includes questions on mood, feelings of guilt, suicide ideation, insomnia, agitation or retardation, anxiety, weight loss, and somatic symptoms. Seventeen items are rated.

A score of 0 to 7 is considered to be normal, 14 to 18 moderate depression, 19 to 22 severe depression and 23 or more very severe depression. This can be helpful for monitoring progress but is only used as a guide.



Expert commentary

This review confirms the view that for moderately severe depression in middle aged patients, there is little to choose between cognitive behavioural therapy and second-generation antidepressants over the first 3-4 months of treatment in terms of either treatment response or adverse effects, including discontinuation. Combining treatment modalities appears to make little difference at this early stage. The evidence regarding this comparison is relatively weak, and plagued as always by the likelihood of high rates of spontaneous remission.  Further research is needed, not least to establish longer term outcomes in rigorous head-to-head trials, although the costs of this are likely to be prohibitive.

Professor Scott Weich, Professor of Psychiatry, Warwick Medical School

Back to top