It’s been a rough few months for people taking antidepressants. We’ve been bombarded with information that insists our medication is ineffective and harmful, that any benefit we gain from taking these pills is simply a placebo effect, and that by accepting a prescription of antidepressants we are joining an ever-growing zombified mass of morons.
From Peter Gøtzsche’s ‘evidence’ that antidepressants don’t just have side-effects but actually kill the people who take them, to the BBC Panorama programme “A Prescription for Murder?” which linked taking antidepressants to violent crime, to Johan Hari’s recent book questioning everything about depression “Lost Connections”, the anti-antidepressants voices have really hit the mainstream.
Of course this is all unsurprising, with around 350 million people worldwide living with clinical depression and NHS prescriptions for antidepressants at an all time high, there is clearly something to be concerned about. Are we medicalising normal human experience, paying too much heed to the pharma-marketing and overprescribing powerful psychotropic medication, or are we seeing increasing levels of illness worldwide that reflect a genuine need for evidence-based interventions (talking therapies, medication and other psychosocial treatments)?
The economic burden of depression in the USA alone has been estimated to be more than US$210 billion. There are lots of evidence-based treatments but because of inadequate resources, antidepressants are used more frequently than psychological interventions. There is considerable debate, both in the scientific literature and in the mainstream media about the efficacy and safety of antidepressants (Ioannidis, 2008), so a major new review in this field is very welcome.
This network meta-analysis (Cipriani et al, 2018) contains the largest amount of unpublished data to date, and open science fans will be delighted to hear that all the data from the study have been made freely available on Mendeley Data. Overall, 522 double-blind RCTs (1979-2016) of 21 commonly used antidepressants were included, which makes this the largest ever meta-analysis in psychiatry and probably the largest network meta-analysis in medicine in terms of number of double blind RCTs included.
The analysis has a total of 116,477 participants (87,052 people randomly assigned to receive a drug, and 29,425 to receive placebo). The majority of patients had moderate-to-severe depression. This review is an update of a previous meta-analysis (Cipriani et al, 2009), which just included head-to-head comparisons of 12 antidepressants, so whichever way you look at it, this is a big piece of important evidence.
RCTs were excluded if they had patients with bipolar depression, symptoms of psychosis or treatment resistant depression.
The authors searched a comprehensive set of databases, the websites of regulatory agencies, and international registers for published and unpublished double-blind RCTs, from inception to Jan 8, 2016. They identified all double-blind RCTs comparing antidepressants with placebo, or with another antidepressants (head-to-head trials) for the acute treatment (8 weeks) of major depression in adults aged 18 years or more. Importantly, they also contacted pharmaceutical companies, original study authors and regulatory agencies to supplement incomplete reports of the original papers, or provide data for unpublished studies. The value of this comprehensive methodology is borne out when you read that they retrieved unpublished information for over half (52%) of the trials included in the meta-analysis (#AllTrials anyone?). Clearly there’s always more unpublished data out there, but the reviewers should be commended for putting together such a comprehensive data-set.
This substantially bigger data-set has given the reviewers the opportunity to investigate outcomes they couldn’t include in their previous review (Cipriani et al, 2009), such as remission, change in mood symptoms and dropouts due to side-effects, as well as a number of methodological issues, such as sponsorship, dosing schedule, study precision, and novelty effect.
- Efficacy (the number of patients who responded to treatment, i.e. who had a reduction in depressive symptoms of 50% or more on a validated rating scale over 8 weeks)
- Acceptability (proportion of patients who withdrew from the study for any reason by week 8).
Unfortunately the reviewers did not have access to individual-level data, so were not able look at the effectiveness or acceptability of antidepressants in relation to age, sex, severity of symptoms, duration of illness or other individual-level characteristics.
- All 21 antidepressants were more effective than placebo
- The most effective antidepressants were agomelatine, amitriptyline, escitalopram, mirtazapine, paroxetine, venlafaxine and vortioxetine
- The least effective antidepressants were fluoxetine, fluvoxamine, reboxetine and trazodone.
- Only clomipramine was less acceptable than placebo
- The most tolerable antidepressants were agomelatine, citalopram, escitalopram, fluoxetine, sertraline and vortioxetine
- The least tolerable antidepressants were amitriptyline, clomipramine, duloxetine, fluvoxamine, reboxetine, trazodone and venlafaxine
The included data covers 8-weeks of treatment, so may not necessarily apply to longer term antidepressant use.
The differences in efficacy and acceptability between different antidepressants were smaller when data from placebo-controlled trials were also considered.
Industry funding was not associated with substantial differences in response or dropout rates, but non-industry funded trials were few and many trials did not report or disclose their funding.
Antidepressants tended to show a better efficacy profile when they were novel and used as experimental treatments than when they had become old. This novelty effect might arise where a novel agent is perceived to be more effective and better tolerated; alternatively, selective analyses and outcome reporting bias might be more prominent when a treatment is first launched.
The reviewers concluded:
These results should serve evidence-based practice and inform patients, physicians, guideline developers, and policy makers on the relative merits of the different antidepressants.
Strengths and limitations
- The design of this network meta-analysis and inclusion of unpublished data is intended to reduce the impact of individual study bias as much as possible
- Of course, the vast majority of included trials (78%) were funded by pharmaceutical companies
- 9% of the included trials were rated as high risk of bias, 73% as moderate and 18% as low
- The quality of many comparisons was assessed as low or very low for amitriptyline, bupropion, and venlafaxine, whereas it was often rated as moderate for agomelatine, escitalopram, and mirtazapine
- Many trials did not report adequate information about randomisation and allocation concealment, which restricts the interpretation of these results
- The lack of individual-level data makes it impossible to investigate issues such as age, sex, symptom severity or duration of illness
- The review did not cover specific adverse events, withdrawal symptoms, or antidepressants used in combination with other non-drug treatments, which is clearly practically useful information for most patients
- The included trials were short (8 weeks) and some antidepressant side effects occur over a prolonged period, so positive results should be interpreted with caution
- This network meta-analysis looks at antidepressants when prescribed as an initial treatment. Given the modest effect sizes, non-response to antidepressants will occur. These results are not relevant to treatment-resistant depression and the reviewers point out that evidence in this group of patients is much more scarce
- The study was funded by the NIHR Oxford Health Biomedical Research Centre and the Japan Society for the Promotion of Science.
Implications for practice
The review strikes a nice balance between presenting this strong evidence that antidepressants work for adult depression and accepting the limitations and potential biases in this data. Prescribers should familiarise themselves with the detail in this open access paper and appendix, as it contains a huge amount of useful data for patients, clinicians and guideline developers.
We know from another recent meta-analysis by Cipriani et al (2016) that fluoxetine is probably the only antidepressant that might reduce depressive symptoms in children and adolescents. There are significantly fewer studies in young people so much more uncertainty about the risks and benefits of antidepressant treatment.
The reviewers are optimistic about how future research could also include the individual-patient data missing from this analysis, so-called individual-patient data network meta-analysis:
This analysis will allow the prediction of personalised clinical outcomes, such as early response or specific side-effects, and the estimate of comparative efficacy at multiple timepoints.
So, I’m going to carry on taking my antidepressants, having read this new evidence. I already knew that they worked for me, along with the talking treatment, exercise, diet, mindfulness and music that I’ve been using in combination since I was diagnosed a few years ago.
I hope this evidence offers some reassurance to other people who take medication for their depression. Of course drugs are not the only treatment, but they are a vital part of the range of interventions we have to help people live productive and fulfilling lives.
Cipriani A, Furukawa TA, Salanti G, et al. (2018) Comparative efficacy and acceptability of 21 antidepressant drugs for the acute treatment of adults with major depressive disorder: a systematic review and network meta-analysis. Lancet 2018; published online Feb 21. http://dx.doi.org/10.1016/S0140-6736(17)32802-7.
Ioannidis JP. (2008) Effectiveness of antidepressants: an evidence myth constructed from a thousand randomized trials?
Philos Ethics Humanit Med 2008; 3: 14.
Cipriani A, Furukawa TA, Salanti G, et al. (2009) Comparative efficacy and acceptability of 12 new-generation antidepressants: a multiple-treatment meta-analysis. Lancet 2009; 373: 746–58.
Cipriani, A et al. (2016) Comparative efficacy and tolerability of antidepressants for major depressive disorder in children and adolescents: a network meta-analysis. The Lancet , Volume 388 , Issue 10047 , 881 – 890.