From the Editor

Last week, I spoke to a patient about antidepressants. “But do they really work?” she asked. While antidepressants are commonly prescribed, many patients wonder about them. That’s not surprising: in popular culture, these medications are often portrayed as risky and unhelpful. Just a few weeks ago, the most popular women’s fitness magazine in the world described fawningly how a woman quit her medications and felt better (“I felt more alive and in control of my emotions with each passing day”). A few years ago, a major study suggested that antidepressants basically match placebo in efficacy; 60 Minutes covered it.

And now there is the new Cipriani et al. paper. “We found that all antidepressants included in the meta-analysis were more efficacious than placebo in adults with major depressive disorder…”

Is this the biggest psychiatry paper of the year? Certainly, it may be one of the most impressive. It took six years of effort. Oxford University’s Andrea Cipriani and his co-authors drew from all available data – published and unpublished, and covering more than 500 trials.

The media coverage has been incredible. The Guardian summarized the paper with the first two words of its article: “Antidepressants work.”


In this Reading, we look at the big study and mull the big implications.



Antidepressants and Placebo

“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”

Andrea Cipriani, Toshi A Furukawa, Georgia Salanti, Anna Chaimani, Lauren Z Atkinson, Yusuke Ogawa, Stefan Leucht, Henricus G Ruhe, Erick H Turner, Julian P T Higgins, Matthias Egger, Nozomi Takeshima, Yu Hayasaka, Hissei Imai, Kiyomi Shinohara, Aran Tajika, John P A Ioannidis, John R Geddes

The Lancet, 21 February 2018 Online First (Open Access)

Psychiatric disorders account for 22·8% of the global burden of diseases. The leading cause of this disability is depression, which has substantially increased since 1990, largely driven by population growth and ageing. With an estimated 350 million people affected globally, the economic burden of depressive disorders in the USA alone has been estimated to be more than US$210 billion, with approximately 45% attributable to direct costs, 5% to suicide-related costs, and 50% to workplace costs. This trend poses a substantial challenge for health systems in both developed and developing countries, with the need to treat patients, optimise resources, and improve overall health care in mental health.

Grouped into various classes of drugs with slightly different mechanisms of action, antidepressants are widely used treatments for major depressive disorder, which are available worldwide. However, there is a long-lasting debate and concern about their efficacy and effectiveness, because short-term benefits are, on average, modest; and because long-term balance of benefits and harms is often understudied. Therefore, innovation in psychopharmacology is of crucial importance, but the identification of new molecular targets is difficult, primarily because of the paucity of knowledge about how antidepressants work. In routine practice, clinicians have a wide choice of individual drugs and they need good evidence to make the best choice for each individual patient. Network meta-analyses of existing datasets make it possible to estimate comparative efficacy, summarise and interpret the wider picture of the evidence base, and to understand the relative merits of the multiple interventions. Therefore, in this study, we aimed to do a systematic review and network meta-analysis to inform clinical practice by comparing different antidepressants for the acute treatment of adults with unipolar major depressive disorder.

9swev4p3Andrea Cipriani

So begins a major new paper by Cipriani et al.

Here’s what they did:

  • The authors did “a systematic review and network meta-analysis.”
  • They began by searching various databases with broad search terms (like “depress*”).
  • They included double-blind, randomized controlled trials (RCTs) comparing antidepressants with placebo or another active antidepressant for depression. Inclusion criteria: patients over 18 with a diagnosis of major depressive disorder.
  • They included all second-generation antidepressants, as well as 2 older ones (amitriptyline and clomipramine), and trazodone and nefazodone.
  • They then did manual searches for “published, unpublished, and ongoing RCTs in international trial registers, websites of drug approval agencies, and key scientific journals in the field…” The authors then reviewed all the papers found. (!)
  • “Our primary outcomes were efficacy (response rate measured by the total number of patients who had a reduction of ≥50% of the total score on a standardised observer-rating scale for depression) and acceptability (treatment discontinuation measured by the proportion of patients who withdrew for any reason).”
  • Complex statistical analyses were done, beginning with the authors estimating odds ratios and standardized mean differences. “The study effect sizes were then synthesised using a random-effects network meta-analysis model.”

Here’s what they found:

  • The analysis has a total of 116,477 participants (87,052 people randomly assigned to receive a drug, and 29,425 to receive placebo) in 522 double-blind RCTs – “421 trials from the database search, 86 unpublished studies from trial registries and pharmaceutical company websites, and 15 from personal communication or hand-searching other review articles.”
  • Demographically: The mean age was 44 years; 62.3% were women. Geographically: 48% of recruited patients were from North America; 7%, from Asia; 27%, from Europe. And in terms of illness experience: “the great majority of patients had moderate to severe depression.”
  • “In terms of efficacy (432 RCTs, comprising 102 443 patients), all antidepressants were more effective than placebo, with ORs ranging between 2·13 for amitriptyline and 1·37 for reboxetine.”
  • “In terms of acceptability (422 RCTs, comprising 99 787 patients), agomelatine (OR 0·84) and fluoxetine (0·88) were associated with fewer dropouts than placebo; by contrast, clomipramine was worse than placebo (1·30).” See the figure below.
  • Considering the head-to-head studies and effectiveness: The most effective antidepressants were agomelatine, amitriptyline, escitalopram, mirtazapine, paroxetine, venlafaxine and vortioxetine. The least effective antidepressants: fluoxetine, fluvoxamine, reboxetine and trazodone.
  • Considering the head-to-head studies and tolerability: The most tolerable antidepressants were agomelatine, citalopram, escitalopram, fluoxetine, sertraline and vortioxetine. The least tolerable antidepressants: amitriptyline, clomipramine, duloxetine, fluvoxamine, reboxetine, trazodone and venlafaxine.


They write:

We found that all antidepressants included in the meta-analysis were more efficacious than placebo in adults with major depressive disorder and the summary effect sizes were mostly modest. Some antidepressants, such as escitalopram, mirtazapine, paroxetine, agomelatine, and sertraline had a relatively higher response and lower dropout rate than the other antidepressants. By contrast, reboxetine, trazodone, and fluvoxamine were associated with generally inferior efficacy and acceptability profiles compared with the other antidepressants, making them less favourable options. To make our results as relevant and robust as possible to inform clinical practice, we decided to focus on head-to-head studies and at the same time emphasise the certainty of the retrieved evidence. Our assessment overall found few differences between antidepressants when all data were considered, while there was more diversity in the range of efficacy and dropout patterns seen across the head-to-head comparisons than the meta-analysis of antidepressants versus placebo.

The paper is accompanied by a Comment by Sagar V. Parikh and Sidney H. Kennedy, “More data, more answers: picking the optimal antidepressant.

You can find it here:

It begins:

In an era of increasingly large datasets for health and emphasis on so-called big data analyses, key clinical questions remain unpretentiously simple. For example, do some antidepressants work better than others for depression? And are some more tolerable than others, at least as measured in dropout rates? A quick PubMed search of antidepressant meta-analyses yields more than 2000 hits, but the complexity of understanding which antidepressants are better or more tolerable than others is made particularly daunting by the fact that more than 40 antidepressants are available.

parikh-headshotSagar V. Parikh

They comment on the rigor of the work:

In this latest paper, Cipriani and colleagues carefully follow recommended procedures to optimise methodological rigour and identify potential sources of bias and error. They also sought to maximise clinical relevance by focusing not only on modern antidepressants but also including the two WHO recommended essential antidepressants, amitriptyline and clomipramine. Special effort to minimise bias was achieved by obtaining additional unpublished data for more than half of the studies, and also by separate analyses to look for various potential contributors to bias including pharmaceutical sponsorship. Finally, for clinical relevance and face validity, although network meta-analysis was used to generate overall findings, results from head-to-head clinical studies were emphasised.

And conclude: “Cipriani and colleagues have made a major contribution.”

A few thoughts…

Size. This is a really big meta-analysis – perhaps the biggest in history – involving more than 500 trials, and almost 120,000 people. As one blogger observed: that’s more people than can fit into Camp Nou, the home stadium of FC Barcelona. And, in the end, the big meta-analysis produced a result that seems right: antidepressants were more efficacious than placebo, but less well tolerated. (To steal a line from University of Toronto’s Benoit Mulsant: I like data that confirm my beliefs.)

3725401Camp Rou

Use of Unpublished Data. The paper drew from published data, yes, but also unpublished data, including drug company data not previously released. For all the talk of secret data and companies withholding, there is no conspiracy – the meds work. But did industry bias taint the results? Cipriani et al. speak directly to this: “In our analyses, funding by industry was not associated with substantial differences in terms of response or dropout rates.”

Limits of Meds. Yes, the meds work – but not incredibly well. The effect size – a measure of their effectiveness – is “mostly modest.” So meds are part of the solution. Let’s not forget, though, about CBT and other forms of evidence-based psychotherapy. This paper confirms the importance of medication management – but also the importance of more than medication management.

Big isn’t Perfect. There are numerous limitations to this important study. To name a few: many trials didn’t comment in a comprehensive way on randomization; older medications (including the MAOIs) weren’t included; the study looked at efficacy at eight weeks – hardly a full picture given the chronic nature of this illness.

So, have Cipriani et al. come up with a definitive list of antidepressants for clinical use? They did find small statistical differences between antidepressants – though methodology was a factor. (As Dr. Mulsant pointed out to me, they write specifically on this in the Discussion: “Our assessment overall found few differences between antidepressants when all data were considered, while there was more diversity in the range of efficacy and dropout patterns seen across the head-to-head comparisons than the meta-analysis of antidepressants versus placebo.”) But does that translate into clinical differences relevant for your patients? You can read the paper and judge for yourself.

Will this study change much? For clinicians like me, antidepressants remain an important part of treatment for depression. But for people like the patient I saw last week, this paper and its positive press may help make the decision to start a medication trial a bit easier to make.

Cipriani et al.’s study may or may not be the biggest psychiatric paper of the year. It is, though, an important public health service.

(Many thanks to Dr. Benoit Mulsant for helping me understand this paper.)


Reading of the Week. Every week I pick articles and papers from the world of Psychiatry.