From the Editor
Antidepressants don’t work. Medications fail to address the real cause of depression. ECT is basically a placebo.
These statements are controversial, but they are mentioned often – including by some of our patients. But what does the literature say about depression management? This week, we look at the debate over antidepressants and ECT, drawing on two recent papers from Psychological Medicine.
In the first selection, John Read (of the University of East London) and Dr. Joanna Moncrieff (of University College) argue that our approach to depression is flawed. In a longer paper that draws on more than 120 references, they challenge basic assumptions about mental health care, arguing against antidepressants and ECT. They advocate for an alternative: “Understanding depression and anxiety as emotional reactions to life circumstances, rather than the manifestations of supposed brain pathology, demands a combination of political action and common sense.”
Were Ali and Frazier having a fight over depression management?
In the other selection, Dr. Carmine M. Pariante (of the King’s College London) agrees to disagree. In a Psychological Medicine paper, he responds. “I have written a piece that tries to put together their point of view with the available evidence, while acknowledging the complexity of the debate.”
Selection 1: “Depression: why drugs and electricity are not the answer”
John Read and Joanna Moncrieff
Psychological Medicine, 1 February 2022 Online First
With the World Health Association and the United Nations calling for a paradigm shift away from the medicalisation of human distress, new evidence about millions of people struggling to get off antidepressants, and ongoing debate about the value and safety of electroconvulsive therapy (ECT), it seems timely to discuss these two longstanding treatments offered to us by biological psychiatry’s ‘medical model’ when we become sad or depressed.
So begins a new paper by Read and Moncrieff.
They focus on several ideas; we outline four:
Are antidepressants active placebos?
“Certain drugs have been referred to as ‘antidepressants’ since the 1950s. Despite this appellation, it is not clear that they have any specific antidepressant effects. Hundreds of placebo-controlled trials suggest that antidepressants are marginally better than placebo at reducing depressive symptoms as measured by depression rating scales. Combining published and unpublished studies suggests an effect size of around 0.3 across different meta-analyses, which translates into a difference of around 2.0 points on the commonly used Hamilton Depression rating scale (HAM-D), which has a maximum score of 52 points. This has not been shown to be a clinically relevant difference. Matching HAM-D scores against Clinical Global Impression scale scores suggests a difference of 8 points is required to indicate ‘mild clinical improvement’ and that a difference of 3 points and below does not register as indicating any change.
“Moreover, the small difference between antidepressant drugs and inert placebo tablets does not confirm that the drugs have an antidepressant action.”
Antidepressants perform poorly in clinical practice
“The majority of placebo-controlled trials have been conducted by the pharmaceutical industry, which has an investment in inflating results, but government-funded research also fails to confirm that antidepressants have beneficial effects. The massive sequenced treatment alternatives to relieve depression (STAR-D) study of gold-standard naturalistic antidepressant treatment produced dismal results. Although the original publication suggested reasonable remission rates of 37% at 12 weeks, the study has been criticised for the use of a secondary outcome measure, exclusion of early dropouts and numerous other protocol deviations. A subsequent analysis found that only 108 participants out of a total of over 4000 recovered, remained well and completed the study, a figure that is still based on the secondary outcome. For 14 years no data was published on the primary outcome measure, the Hamilton rating scale. When independent researchers eventually obtained the data it emerged that people given antidepressant treatment along with high-quality care showed a reduction in their scores of 6.6 points after 12 weeks. This is at the lower end of the range of change seen in people on placebo in meta-analyses of randomised trials and roughly half that in randomised trials comparing different antidepressants. This suggests that the conditions of being in a randomised trial inflate drug effects and that, in real life, antidepressants are no better than placebos.”
Do antidepressants correct an underlying biological abnormality?
“Despite claims by professional organisations and the pharmaceutical industry that depression is due to a chemical imbalance that can be rectified by drugs, there is no evidence that there are any neurochemical abnormalities in people with depression, let alone abnormalities that might cause depression. Where differences between people with and without depression have been found, these are likely to be explained by prior use of antidepressants and other medications; but in most areas, no consistent differences have been found in any case. Although the public, internationally, continue to favour psycho-social explanations of depression, an increasing proportion have been influenced to believe that depression is caused by a chemical imbalance, and across the world increasing numbers now take antidepressants. Seventeen per cent of the population of England were prescribed an antidepressant in 2018, and 14% of US adults by 2015. Yet the number of people requiring services or going onto long-term disability due to depression is increasing.”
“Like antidepressants, the story of ECT is also the story of the power of placebo effects. Positive expectations affect prescribers as well as patients. They influence perceptions of recovery as well as recovery itself. Neurologist John Friedberg pointed out that the rapid spread of ECT across Europe and the USA in the 1940s took place despite there being no studies comparing recipients and non-recipients, and that ‘the influence of ECT was on the minds of the psychiatrists, producing optimism and earlier discharges.’
“The standard placebo in ECT studies, known as ‘sham ECT’ (SECT), is the administration of the general anaesthetic but not the electricity or subsequent convulsion. A review of the literature on placebo responses to ECT concluded: ‘Rigorously defined endogenously depressed patients did exceptionally well with sham ECT, just as well as with real ECT’.
“In the 84 years since the first ECT there have only been 11 randomised placebo-controlled studies (RCTs) for its target diagnosis, depression, all before 1986. A recent review, involving Dr Irving Kirsch, Associate Director of Placebo Studies at Harvard Medical School, highlighted the poor quality of the 11 studies: ‘Only four studies describe their processes of randomisation and testing the blinding. None convincingly demonstrate that they are double-blind. Five selectively report their findings. Only four report any ratings by patients. None assess Quality of Life. The studies are small, involving an average of 37 people.’”
The paper offers an alternative:
“We learn how to support our fellow humans through our life experience, through being cared for ourselves, and sometimes through art and literature. Classifying anxiety, depression and other emotional reactions as mental diseases or disorders obscures the relation between our moods and our circumstances. It leads society to believe that social structures are unchangeable. Instead, we need to listen carefully to the message that people’s emotional reactions convey, and endeavour to create a society in which all people can flourish.”
The paper is clear and well argued. Is it persuasive? Carmine M. Pariante certainly doesn’t feel that it is. He responds with a paper of his own, also published in Psychological Medicine, “Depression is both psychosocial and biological; antidepressants are both effective and in need of improvement; psychiatrists are both caring human beings and doctors who prescribe medications. Can we all agree on this? a commentary on ‘Read & Moncrieff – depression: why drugs and electricity are not the answer.’”
He begins by noting that he was originally a peer reviewer: “When I submitted my (signed) referees’ comments for the editorial on antidepressant and ECT by Read and Moncrieff recently published in Psychological Medicine, I had hoped that there was an opportunity for the authors to incorporate, or at least acknowledge, my comments (and those of fellow referees) into a piece that was less divisive, and more nuanced, than usual.”
In his view, they don’t. Drawing on the Read and Moncrieff paper, he makes a series of arguments; we summarize four:
Yes, ‘antidepressants do not tackle the causes of depression’ – but most of the other medications that we use in medicine also do not tackle the causes of the disorder
“Painkillers take the pain away, not the cause of the pain; anti-hypertensives lower blood pressure, they do not cure hypertension; statins lower cholesterol, they do not cure the genetic problem that generates the high levels. Dexamethasone and heparin save people infected by COVID, even if these drugs do not even remotely affect the virus infection. And so on, and so forth.
“Arguably, only antibiotics or antivirals tackle the cause of a disorder. Or surgery (and not every time).
“Why should medications for depression have a different value threshold? Why should antidepressants only be relevant if they act on the cause of the disorder?”
Yes, there is ‘a small difference’ between antidepressants and placebo using multi-items depression rating scales – but antidepressants specifically, robustly and consistently improve depressed mood
“The Hamilton Depression Rating Scale, still widely used in clinical trials because it is indicated by regulatory agencies, was published before I was born (believe me, that was a long time ago), at a time when the only antidepressants available were tricyclics and (old) MAO inhibitors. The effects of these antidepressants on some of the individual items of the scale (which include many physical symptoms, like sleep, energy, appetite and libido) were different from those induced by the more recent, widely used selective serotonin reuptake inhibitors. Because of this, there is evidence that the total score of this multi-item scale underestimates the efficacy of the newer antidepressants.
“The key question is: are antidepressants, in fact, anti-depressant?
“Yes, they are, because they specifically, robustly and consistently improve depressed mood. The ‘number needed to treat’ (NNT) for antidepressants, a clinically-relevant measure of the effectiveness of medications, is around 7–10, which is similar to medications for other disorders.”
The most recent appraisal of evidence on ECT confirms that ECT results in decreased risk of suicide, improved functional outcomes and quality of life, and decreased rates of rehospitalization
“I am citing from the most recent review on this topic, published only a few days ago in the New England Journal of Medicine.
“It goes on by saying: ‘Trials of ECT for major depressive disorder in patients with treatment-resistant depression have shown pooled response rates of 60% to 80% and pooled remission rates of 50% to 60%’.
“It finishes by saying: ‘Stigma and lack of access to treatment have contributed to the underuse of ECT.’”
Antidepressants are one of the tools to help patients coping with the ‘life circumstances’ and the ‘social conditions’ that causes these ‘emotional reactions’, while we are waiting (and hoping) for society to change
“Imagine that you are a patient with cancer, and that you are told that the reason why you have cancer is a combination of genetic predisposition and of societal factors that are worsened by adverse life circumstances, such as lack of healthy food and high pollution levels in the area where you live now or grew up as a child. This is an accurate statement, and it would be equally accurate for a patient with depression, although we would need to add poverty, discrimination and abuse as additional societal factors.
“Now imagine saying to this patient with cancer that, since the causes of cancer are societal, they should not be receiving any pharmacological treatment, but instead all the efforts should be focussed on improving society.
“And, of course, that they should not trust materialistic oncologists that just want to give them chemotherapy, which does not tackle the societal causes of their cancer.”
A few comments:
1. Both papers are well written.
2. Are John Read and Joanna Moncrieff persuasive? Does Dr. Pariante carry the day? You can read the papers and draw your own conclusions. (For the record, I’m not retiring my prescription pad.)
3. Read and Moncrieff offer good arguments. But do they offer a good alternative by calling for “political action and common sense”?
4. If this debate catches your fancy, others responded to the Read and Moncrieff paper; they also offer a final rebuttal, which can be found here:
The full Psychological Medicine papers can be found here:
Reading of the Week. Every week I pick articles and papers from the world of Psychiatry.