Efficacy of antidepressants
BMJ 2008; 336 doi: https://doi.org/10.1136/bmj.39510.531597.80 (Published 06 March 2008) Cite this as: BMJ 2008;336:516
All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
The editorial by dr. Turner does not address the limitations of using effect sizes (ES) based on sum-scores derived from rating scales such as the Hamilton (HAM-D) or the Montgomery-Åsberg (MADRS) when evaluating efficacy of antidepressants. Those scales contain items that partly are not related to core phenomena in clinical depression. In our own study comparing the antidepressant effect of emotional support by a GP supplemented with antidepressant drugs or placebo (Malt et al, BMJ 1999), the mean difference in ES of adding drug compared to placebo was similar to the ES reported by both Turner et al (New Engl J Med 2008) and Kirsch et al (PlosMed 2008).
However, ES of single items (N=372; intention to treat) revealed significant higher differences in ES compared to the placebo-group for some core symptoms of depression, e.g. insomnia (ES 0.59), pessimistic thoughts (ES 0.54) and reported sadness (ES 0.47)(Malt, Br J Psychiatry 2002). The difference in ES between antidepressants and placebo were much lower for other items such as reduced appetite, observed sadness, concentration problems and suicidal ideation.
Our findings suggest that the mere use of total sumscores of rating scales for calculating differences in ES between treatments - as done when analyzing the FDA-data - may obscure important clinical differences between treatments. Future discussion of the efficacy of antidepressant or psychotherapy for the treatment of mood disorder should take such considerations into account.
Competing interests: I have received payment for lecturing about psychopharmacology from several pharmaceutical companies, Public Health Trusts and organizations. My spouse is currently medical adviser in Pfizer, Norway
Competing interests: No competing interests
The letter from Dr Dermot Walsh reminds me that in the 1970s I compared suicidal overdose deaths from tricyclic antidepressants in the previous decade with those from amphetamines when, as he notes, the latter were commonly prescribed for depression.[1]
Conservatively analysing the annual self-poisoning figures, I found that most years saw 0-2 amphetamine suicides against 2-300 for tricyclics. This important issue only became more widely discussed when the arrival of SSRIs in the late 1980s highlighted tricyclic toxicity. However, given the lack of compelling evidence that SSRIs have specifically reduced suicide rates, it seems quite possible that tricyclics caused more deaths from overdoses than they may have saved from suicide.
I referenced a neglected paper by Leyburn[2], who noted the paradox that since the introduction in the late 1950s, of tricyclic antidepressants, which supposedly nipped depression in the bud, the number of admissions for both moderate and severe depression steadily increased. Changes in admission procedures following the 1959 mental health act reforms (removing the need to be declared insane by magistrates) were not considered a major factor in this increase.
I do not conclude from current and previous debates that pharmacology has no place in treating depression. However, I entered practice at a time when deliberately prescribing placebos, far from being anathema, was regarded as a potentially useful diagnostic exercise. I even administered placebo Electo-Convulsive Therapy (ECT). Two of the three patients made a good recovery, one of them thanking me for the nicest ECT she had experienced. (The third did not improve even when placebo ECT was followed by the real thing.) Few doctors who qualified in the inevitably anti- placebo culture of patients’ rights have any idea of just how powerful non -specific placebo effects can be and many are perennially surprised at their extent and rediscovery.[3,4]
With luck, about 40-50% of depressed (or ‘depressed’) patients respond to the non-specific effects of antidepressant prescribing; an additional 10-20% may respond to their specific pharmacological effects (whatever the alleged mechanisms) and 20–30% do not respond. If these non- responders have ECT, many improve but the large Medical Research Council trial[5] found that the advantage of real over placebo ECT (ie anaesthesia, attention, anticipation) was modest overall. The most seriously depressed patient, whose inclusion caused much ethical discomfort, recovered but was found to have been in the placebo arm. Unfortunately, doctors are probably stuck with current anti-placebo attitudes, at least in societies well-supplied with lawyers and investigative journalists. Of course, these institutions, like patients’ rights, are useful and important but like several other useful developments, they come at a cost. Placebos now flourish only among ‘alternative’ practitioners. They, too, typically underestimate their potency. The late Peter Skrabanek apparently believed that for maximal placebo effect, both prescriber and patient had to believe in it.[6] The evidence strongly suggests that we would be foolish not to.
REFERENCES
1. Brewer C. Suicide with tricyclic antidepressants. BMJ 1976;2:110
2. Leyburn P. A critical look at antidepressant drug trials. Lancet 1967, 2(7526), 1135-8.
3. Shapiro A, Shapiro E. The powerful placebo: from ancient priest to modern physician. 1997; Baltimore,. Johns Hopkins.
4. Andrews G. Placebo response in depression: bane of research, boon to therapy. Br J Psychiat. 2001 Mar;178:192-4.
5. Johnstone E, Deakin J, Lawler P, Crow T. et al. The Northwick Park electroconvulsive therapy trial. Lancet 1980; 1317-20.
6. Steer P. Medical Classics. Follies and Fallacies in Medicine by P Skrabanek and J McCormick. BMJ.2008. 336. 673 (Book downloadable at www.medicine.tccl.ie/public_health_primary_care/skrabanek/publications.php)
Competing interests: None declared
Competing interests: No competing interests
I think David Kernick is being very polite and kind to academics and researchers. They have not done terribly well over these last few decades and I suspect if I were a 'GP' I too would have ceased paying much attention to their scribblings long ago.
Competing interests: None declared
Competing interests: No competing interests
Apologies, depression rating scales are ordinal, not interval scales.
The sentence "However, the HAM-D (and others like it) is an interval scale - higher scores mean more severe levels of depression, but that is all we can conclude." should instead have read "is an ordinal scale".
Competing interests: None declared
Competing interests: No competing interests
Tanu Pramanik.BSc.MA(Sociology)Medical Sociologist and Prof(Dr)Jogenananda pramanik.MD faculty of Medicine,St.Martinus University,Ottrobanda, Curacao,Netherlands Antilles. Correspondence:E-mail:j.pramanik@martinus.edu
Use/misuse/abuse of antidepressants is a matter of serious concern in current medical practice.We read this insightful editorial with interest and applaude the editors view point(1).Hypothetically, if antidepressants are not worth taking, then what should doctors and patients do? Kirsch and colleagues recommend that if antidepressants are to be used at all they should be used only when alternative treatments have failed to provide a benefit(2).Current clinical recommendation is that when considering the potential benefits of treatment with antidepressants, be circumspect but not dismissive. Large number of antidepressant medicines are available in the market with uncertain efficacy.Therefore a stringent mechanism is urgently required to screen those products.On the otherhand,we need to advocate for alternative methodologis to alliviate depression before prescribing antidepressants.Habit forming antidepressants are causing more harm to our society than others.There are several sub-clinical cases of depression patients in our modern society (3)who lead their life without antidepreesant medicines.There are several methods available to handle depression.Our medical professionals need to be properly oriented and respectful to depression patients.During our teaching and training in medical Universities,we need to emphasize the importance of indifying depression patients at the early stages and take care of them.Medical students need necessary exposure to this area and develop clinical acumen to diagnose the vicious cycle of depressive factors working on potetially vulnerrable groups of people in our society (4). This editorial may be an eye-opener for the govermental and other regulatory agencies operating in different countries to intensify their vigil. References: 1.Editorial:Efficacy of antidepressants BMJ 2008;336:516-517 (8 March), doi:10.1136/bmj.39510.531597.80 (published 4 March 2008) 2.Kirsch I, Deacon BJ, Huedo-Medina TB, Scoboria A, Moore TJ, Johnson BT. Initial severity and antidepressant benefits: a meta-analysis of data submitted to the Food and Drug Administration. PLoS Med 2008;5:e45. 3.News:Depression affects elderly people's lives more than physical illnesses;Janice Hopkins Tanne BMJ 2004;329:1307 (4 December), doi:10.1136/bmj.329.7478.1307-c 4.Medical students knowledge and attitudes toward senior citizens and aged patients:Prof (Dr)Jogenananda Pramanik.MD,Professor,Dept of Pathology St James School of Medicine,4 Krelendijk,Bonaire Netherlands Antilles.,Tanu Pramanik,BSc,MA.Medical Sociologist.http://www.bmj.com/cgi/eletters/329/7478/1307-c#123057. -----------------------------------------------------------
Competing interests: Authors worked in elderly homes in India,Nepal,China and caribbean Islands.
Competing interests: No competing interests
The current debate over the relevance of antidepressants - the interpretation of meta-analysis, what is clinically significant, uncertainty over major side effects and the confounding influence publication bias, clearly illuminates why GPs are indifferent to research based evidence. I have been prescribing SSRIs for the last 20 years. Psychological interventions are essentially unavailable in the NHS, the medication is cheap and gives few side effects, I don’t have access to a placebo and patients invariably come back saying they feel better. That’s good enough for me.
Yours faithfully David Kernick
Competing interests: None declared
Competing interests: No competing interests
One factor that seems to have been overlooked in this discussion is whether part of what we are seeing here - the relatively greater efficacy of anti-depressants in the more severely depressed - is simply an expression of the law of diminishing marginal returns. The very severely depressed have a long way to go on the climb back down the 'HAM-D' (Hamilton Depression Rating Scale) ladder and often very small changes in one biological factor (such as sleep or appetite) can have substantial knock-on effects on other symptoms, thereby assisting the individual down several rungs simultaneously. The same degree of gain becomes more difficult to achieve as time goes on or if initial symptom levels are not as high. There may not be as much benefit to be had from the biological changes that antidepressants may bring about when initial symptoms are lower, or in any given individual over time when the more 'stubborn' and trait-like aspects of the illness are all that remain. If this is true, it would be no surprise that the effect size is smaller in those with less severe depressive symptoms.
Another aspect that has rarely been appreciated is the appropriateness of treating scales such as the HAM-D as if they are ratio scales, i.e. as if every single-point reduction means the same thing regardless whether it appears in the context of a high or low score. However, the HAM-D (and others like it) is an interval scale - higher scores mean more severe levels of depression, but that is all we can conclude. We cannot say that a person who scores 20 is twice as depressed as a person who scores 10. This becomes a very important point when the discussion boils down to whether an effect size represents a clincally meaningful change. A 2-point reduction in the HAM-D could mean something very different to a person (and to the friends, relatives and carers who are worrying about the person's safety) if it takes them from 'Wishes he were dead' to 'No thoughts of death or suicide', compared to if it takes them from 'Worries about minor matters' to 'No anxiety'. If a small effect size means only a few points change in HAM-D score, unfortunately we can conclude very little about the meaningfulness of that change through merely knowing its magnitude.
Naturally, in order to make illnesses such as depression more amenable to research, scales such as this are necessary. But it is essential to remember exactly what we have done in constructing such a scale - put numbers on feelings, emotions and bodily states. But just because we have made them look like numbers, does not mean that they will behave like them.
Competing interests: None declared
Competing interests: No competing interests
I wish to correct a typographical error that appears approximately two-thirds of the way into our editorial. The parenthetical phrase that reads, “defined arbitrarily as P≥0.5” should read “defined arbitrarily as d≥0.5”.
Competing interests: Following a query from the editor about conflicts of interest I append the following information, which has been disclosed in previously published articles. From 1998 to 2001, I served as a medical reviewer at the US Food and Drug Administration (FDA). Subsequently, but ending in 2005, I provided outside consulting to Bristol-Myers Squibb, Eli Lilly, GlaxoSmithKline. From 2004 to 2005, I was on the speaker's bureaus of Eli Lilly, AstraZeneca, and Bristol-Myers Squibb.
Competing interests: No competing interests
NEUROPHARMACOLOGICAL THERAPY OF ENDOGENOUS DEPRESSION Fuad Lechin, MD, PhD; Bertha van der Dijs, MD Universidad Central de Venezuela flechin@telcel.net.ve
We will refer to the article by Turner and Roshental (1) dealing with the efficacy of antidepressants. We would like to inform that we demonstrated in 1995 that endogenous depression (ED) is underlied by raised plasma levels of noradrenaline (NA) plus lowered plasma levels of adrenaline (Ad) (2, 3). Our findings have been ratified by many other researchers. In addition to the above, it should be known that NA plasma levels are positively correlated with the activity of the A5-NA pontine nucleus, responsible for the peripheral neural sympathetic activity whereas Ad plasma levels (adrenal sympathetic activity) are positively correlated with the adrenal sympathetic activity. This peripheral sympathetic branch depends on the excitation of the C1-Ad medullary nuclei. Furthermore, both the A5-NA and the C1-Ad nuclei interchange inhibitory axons in such a way that Ad and NA released from both types of central nervous system (CNS) nuclei provoke the inhibition of both targets by acting at pre-synaptic alpha-2 receptors which crowd the above mentioned nuclei (4, 5). In addition to the above, we also demonstrated that patients affected by ED presented low levels of plasma tryptophane (trp) and raised platelet serotonin (p-5HT) levels. With respect to the latter, exhaustive evidence have demonstrated that these peripheral parameters are positively correlated with the activity of dorsal raphe (DR) and median raphe (MR) serotonergic nuclei, respectively (6). Thus, we postulated that endogenous depression presents with a MR-5HT over DR-5HT predominance plus the overactivity of the A5-NA and the hypoactivity of the A6-NA + C1-Ad (4, 5).
The above postulation is reinforced by the successful neuropharmacological treatment addressed to revert the above CNS disorder. This target is reached by the administration of small morning dose of an alpha-2 antagonist (yohimbine) plus a NA-uptake inhibitor (desipramine, maprotyline, reboxethine, etc) plus a 5HT-1A antagonist (pindolol, which enhances the firing activity of both the DR-5HT plus the C1-Ad nuclei), and the nocturnal prescription of a small dose of a 5HT-uptake inhibitor (paroxetine, sertraline, fluvoxamine, etc). The addition of a small dose of mirtazapine (alpha-2 + 5HT2 antagonist), favors the release of NA and 5 -HT from the A6-NA or locus coeruleus plus the DR-5HT nuclei, respectively This therapeutical approach should be maintained during 3 to 4 months, at which period should be interrupted (5).
References 1. Turner EH, Rosenthal R. Efficacy of antidepressants. BMJ 2008; 336(7643):516-517.
2. Lechin F, van der Dijs B, Orozco B, Lechin ME, Acosta E, Lechin AE, Baez S, Rada I, Arocha L, Leon G, Garcia Z. Plasma neurotransmitters, blood pressure and heart rate during supine-resting, orthostasis and moderate exercise conditions in major depressed patients. Biol Psychiatry 1995;38:166-173.
3. Lechin F, van der Dijs B, Benaim M. Stress versus depression. Review. Prog Neuro-Psychopharmacol Biol Psychiatry 1996;20:899-950.
4. Lechin F, van der Dijs B, Lechin ME. Neurocircuitry and Neuroautonomic Disorders: Reviews and Strategies of Therapy. Basel: Karger AG, 2002.
5. Lechin F, van der Dijs B. Crosstalk between the autonomic nervous system and the central nervous system: Mechanistic and therapeutic considerations for neuronal, immune, vascular, and somatic based diseases. In: Maiese K, editor. Neurovascular Medicine: Pursuing Cellular Longevity for Healthy Aging. New York: Oxford University Press, 2008 (In Press).
6. Lechin F, van der Dijs B, Hernandez-Adrian G. Dorsal Raphe (DR) vs. Median Raphe (MR) serotonergic antagonism. Anatomical, physiological, behavioral, neuroendocrinological, neuropharmacological and clinical evidences: Relevance for neuropharmacological therapy. Prog Neuro- Psychopharmacol Biol Psychiatry 2006;30(4):565-585.
Competing interests: None declared
Competing interests: No competing interests
Alternatives to Prescription Drugs
As part of a GP surgery in Nottingham, I run a Personal Coaching surgery based on Neuro-Linguistic Programming techniques for those patients who would normally have their antidepressants continually prescribed. The simple fact that this Talking Therapy focuses on the individual and their issues where a doctor is limited by time has an enormous effect on the mental health of the patient.
I have proved over the last 8 months that affecting a person's self-esteem and emotional state is far more rewarding for them as individuals than it is to prescribe a drug to chemically alter their state.
This is a much better approach for those patients who are not extremely depressed and who need a little nudge on their journey. I would urge all GP's to start thinking outside the box on this one as an hour with me is a lot cheaper than a patient coming back over a number of years, but more importantly, it's better for the patient too.
Competing interests: antidepressant use in comparison to NLP-based Coaching
Competing interests: No competing interests