Gail Bell writes evocatively and well. Worryingly well. As an “opinion piece” her essay advances valid concerns about the current use and application of antidepressant drugs in our society. But it is also an “edgy” essay and the edginess makes it difficult to make this a frisson-free response. Let’s look at the essayist’s style before addressing issues of substance, for the question is whether such polemic dressed in the cloak of reason leads to unbalanced conclusions.
Bell’s essay starts with a vignette of Angie. It is an amorphous vignette, so we do not know whether Angie has a primary mood disorder, another psychiatric condition, is suffering from a mix of illicit and prescribed medication, is experiencing a side effect of prescribed medication or is merely “overwrought”. In the absence of such information, any response by a health professional to the question of whether she should stay on antidepressant medication is straightforward. But Bell’s own interpretation of her response is telling – she says she provides “the tame dove answers of my white-coat training”. Commonsense and prudence alone would suggest that she encourage Angie to be reviewed by her health practitioner – as she did – but Bell’s “tame dove” reference suggests, also, a frustration with an assumed health hierarchy and a wider dissatisfaction with the management of depression by health professionals. While health professionals can defend themselves against her anti-allopathic rhetoric, the harm lies in the fact that many depressed patients will be unsettled by her essay.
While I share some of Bell’s concerns, I would like to offer another perspective on several issues. First, some history. As Bell notes, there has been a longstanding division of depressive disorders into two principal “types”. The first, once called “endogenous depression”, now “melancholic depression”, is the quintessentially biological type, with a strong genetic contribution, underpinned by a number of derangements in biological functioning and having some stereotypical features. While minimally responsive to placebo and talking therapies (psychotherapy and counselling), it is highly responsive to physical treatments, including antidepressant drugs.
The second type, once variably called “exogenous”, “neurotic” or “reactive”, was more likely to be brought on by the impact of a stressful life event perturbing an individual with certain vulnerability factors (e.g. personality style), and therefore was best addressed by psychotherapeutic strategies.
This simple binary model – for both modelling and treating depressive conditions – made decisions about antidepressant drug therapy relatively simple. Argued since biblical times, it held sway until the middle of the twentieth century, but lost currency as a consequence of several key changes that, in turn, progressively clouded our capacity to understand “depression”.
The first antidepressant drug (the tricyclic drug Imipramine) was “discovered” a little more than fifty years ago. The manufacturers, Ciba-Geigy, did not wish to take that drug to market as their analyses indicated that there were insufficient depressed people in the world for the drug to return a profit, and it was only after strong protest advocacy in the United States (by consumers) that it was released. When we consider the sales of antidepressants over the last decade, that judgment by Ciba-Geigy may seem inexplicable. But “depression” in the middle of the twentieth century essentially comprised severe expressions of “biological depression” (psychotic or melancholic depression) that resulted in a percentage of people being hospitalised, generally in asylums as few general hospital psychiatry units existed. Such connotations meant that “depression” joined with other severe mental illnesses in being stigmatised, discouraging the many with “less severe” expressions from attending health practitioners. Diagnosis and detection were therefore limited.
Now, for the first time, there were drugs that were strikingly effective in assisting those with melancholic depression. And in medicine, when any therapy is discovered to be useful for a target condition, there is a rapid move to test its efficacy beyond its initial boundaries. Such diffusion compromises clarity, even when the “disease” model is well-defined, but in this case the phenomenon was also wedded to a new “muddy” definition of depression.
The template for modelling and treating “depression” was radically changed by the introduction of a North American model for categorising psychiatric disorders. The Diagnostic and Statistical Manual of Mental Disorders – Third Edition (DSM-III) was published in 1980 by the American Psychiatric Association. In the preceding decade, senior American academic psychiatrists had become increasingly concerned about the psychoanalytic emphasis in North American theorising and practice. They could see psychiatry losing its roots in medicine and risking irrelevance. They sought to bring psychiatry back into medicine by emphasising psychiatry’s biological focus and instituting a strong scientific approach. DSM-III categorisation involved “structured criteria” for disorders (to ensure scientific reliability) but unfortunately ignored any consideration of cause or aetiology.
The DSM-III model characterised depression as a single disorder that varied in severity, duration and persistence. It was a model that described a unitary concept with various “dimensions”, and therefore introduced pseudo-categories of “major depression” and “minor depression”. Yet while the diagnosis of “major depression” has gravitas, in reality it is non-specific – a “muddy” definition. An individual is required to have a significantly depressed mood for at least two weeks, have some degree of impairment, and several symptoms such as appetite or sleep disturbance, loss of energy and loss of interest and pleasure in usual activities. While the DSM-III description of a characteristic “major depressive episode” largely captures the severe biological melancholic depressive disorders, the clinical criteria listed for “major depression” were structured according to one of the guiding principles of DSM-III – to have such criteria “describe the lowest order of inference necessary to describe the characteristic features of the disorder”. Thus, while an individual with melancholic depression might experience such profound and pathological guilt that they feel it their duty to kill themselves, the relevant DSM-III “guilt” criterion now allowed “thin” symptoms such as “feelings of worthlessness and self-reproach” to build to a new definition of “depression”.
Thus, after 1980 “clinical depression” was re-defined, with the “bar” for meeting a diagnosis set low, resulting in many more people with varying levels of depression being diagnosed as a “clinical case”. “Depression” moved from (predominantly) “disease” status to encompass a range of conditions: diseases, disorders, syndromes, predicaments and possibly extensions of normal mood states – although the “new” definition tended to position all such “conditions” as “disease” states.
The redefinition of clinical depression not only led to much higher diagnosed prevalence rates of current and lifetime depression, but now the majority of identified sufferers had the “less severe” non-melancholic or “atypical” expressions, where the role of the tricyclic antidepressants had been shown to be less useful.
The homogenising DSM-III model had a powerful effect on the approach to treatment. Rather than progressively develop a matrix where identified depressive types were married with preferential specific drug and non-drug treatment strategies, an “integrative” movement occurred. “Integrative” theories conceded that clinical depression could be caused by many factors (e.g. genetic, developmental, psychosocial), but these individual nuances could be ignored as they were all channelled along a “final common pathway” to create “major depression” – a muddily defined pseudo-entity that has become entrenched, homogenising multiple expressions of depression, irrespective of their biological, psychological or social cause.
Predictably, subsequent research into this “entity” has failed to identify consistent causes. Perhaps more importantly, when quite different treatments for “major depression” are compared in randomised controlled trials, all appear comparable in their efficacy. Thus, antidepressants of differing classes tend not to differ from each other or from the major psychotherapies or even from “natural” remedies, such as St John’s Wort, in their efficacy.
We should not be surprised by such an opaque result. Test a treatment non-specifically (i.e. as a universal treatment, relevant across the board) for a non-specific pseudo-entity like major depression, and we should anticipate a non-specific result, with all treatments given a guernsey. In essence, the dodo bird verdict: “All have won and all must have prizes.”
Another effect was the emergence of contending colonisers, each seeking to occupy the new territorial map, and each offering a simple view of the muddy landscape. The majority view (reflecting the “new biological psychiatry” and the influence of the pharmaceutical industry) now interprets major depression as a biological condition reflecting a “chemical imbalance” quite independent of personality, and requiring correction of the perturbed neurotransmitter circuits by the prescription of an antidepressant drug. A contending view comes from psychologists who argue that major depression is the consequence of a faulty “attribution style”, and requires cognitive behaviour therapy. And other professionals put forward their own singular treatment modality as the best means of treating “it”.
In essence, a procrustean model has developed whereby the individual’s condition is fitted to the therapist’s discipline or training rather than the therapy being “fitted” to the characteristics of the individual’s disorder. This is quite at variance with the standard medical model and of great concern. Imagine if you had clinically significant breathlessness (“major breathlessness”). You would not expect a general practitioner to prescribe a treatment merely on the basis of that non-specific diagnosis only. You would expect your health practitioner to identify the cause (e.g. asthma, pneumonia or a pulmonary embolus) and then rationally derive a cause-weighted treatment.
It is held that destigmatising depression (a process which has been extremely successful) requires a simple definition. Characterising depression as an “it”, and encouraging recourse to diagnosis and treatment, is the message that dominates most campaigns. Unfortunately, many health professionals have also adopted a simplistic and erroneous model, one that compromises treatment for many sufferers. It seems to us – at the Black Dog Institute – that it is just as easy to communicate a more realistic “horses for courses” model (i.e. that there are many types of depression, some which respond best to medication, others to certain specific psychotherapies, others to support and counselling). We argue that health professionals should seek to establish and disseminate diagnostic and treatment matrices, as are produced in other areas of medicine (e.g. best treatments for differing cancers, for managing differing types of diabetes).
The redefinition of “depression” to include a majority of non-melancholic conditions coincided with the introduction of the narrow-action SSRI antidepressant drugs – and with many sufferers of these “less biological” depressive disorders finding them beneficial. For the first time in my professional career as a psychiatrist, I encountered patients (and people at social gatherings) talking comfortably about how an SSRI antidepressant had aborted their depression and allowed them to function again. Most had a non-melancholic depressive disorder. The benefits of SSRIs stemmed not only from an antidepressant action but also from an apparent capacity to mute emotional dysregulation and worrying, modifying the response to stress and assisting many out of depression, as well as decreasing the chance of recurrence. It was clear that the new SSRI antidepressants were of benefit for a percentage of people with depressive conditions who – previously reluctant to seek help because of stigma – had suffered in silence and in depression-induced darkness.
Because many of these people appear to be functioning well, they risk invoking the “they should pull their socks up instead of taking medication” response. But we have all seen people with medical conditions (e.g. diabetes, blood pressure, cancer) who appear to be functioning well. Do we challenge their right to take medication and diminish them with ad hominem statements inferring self-indulgence or “spinelessness”? It is on this issue that I part company with Bell, for her views risk activating such responses.
Clearly, the SSRIs were promoted over-enthusiastically by the pharmaceutical companies and by many health professionals. While their side-effect profile is only marginally superior to that of the older antidepressants, we initially misjudged the reality that, like most medications, they do have a number of significant side effects for a percentage of people, and can be less effective than the older broad-action antidepressants for managing the melancholic depressive disorders. Nevertheless, a confluence of factors had set the scene.
“Depression” as redefined had a much higher prevalence than had been previously quantified; its non-specific definition allowed the new antidepressant drugs to be positioned as a universal (“one-size-fits-all”) therapy; and drug therapy was cheaper and easier to roll out than psychotherapy. This model also delivered a much broader market to the pharmaceutical industry. And finally, sufferers of a “chemical imbalance” were not blamed for it, nor did they need to contribute to making a recovery from their depression.
As with the take-up of any effective drug (in any area of medicine), excessive enthusiasm has tempered over time. Steroids are another example of this. Initially perceived as wonder drugs, their extremely troubling side effects for a percentage of people were progressively identified and they were also found to be less effective than first judged. The same phenomenon has occurred with the newer antidepressants, with all the passion of a new relationship and its course reflected in the literature. Kramer’s book Listening to Prozac captured the proselytising evangelical infatuation stage, echoed in initial media reports. Later, the media became less endeared with the SSRIs (a “so what, everyone’s on them” scenario), with Beyond Prozac and Prozac Nation capturing a drift from fidelity. And in recent times, the backlash evidences a falling out of love, with stories emphasising the side effects of the SSRIs, criticising the pharmaceutical industry and sensationalising any real risk of SSRI-induced suicidality – as captured in the book Let Them Eat Prozac. “Divorce” is now in the offing … some antidepressants risk being withdrawn from market or having their prescription markedly curtailed.
It is human nature to respond in anger when we have invested our faith in something and been disappointed when it falls short. The story is a regrettably common one in medicine (e.g. L-Dopa for Parkinson’s Disease; HRT treatment) and therefore not unique to psychiatry. The risk is in throwing the baby out with the bath water.
How to proceed when there is a dumb model for the depressive disorders out there, which homogenises multiple depressive types to “depression as an ‘it’”, which ignores cause, and which encourages a view that treatments are universally relevant? In such circumstances, it must be expected that some individuals will be “under-treated” (e.g. those with a melancholic disorder who fail to receive a physical treatment) and others effectively “over-treated” (e.g. those who neither need nor respond to an antidepressant drug, and who receive a seemingly endless parade of drug therapies). Mismanagement of depression occurs as much from errors of omission (e.g. a therapist giving meandering chicken-soup advice to a sufferer month after month) as from commission (e.g. excessive reliance on medication), and it is the paradigm failures across the board – not just those associated with antidepressant medication – that need to be identified and addressed. As noted earlier, I argue for more horse sense – respecting sub-typing and differential treatment recommendations, a “horses for courses” paradigm that allows situations when an antidepressant (and the right one) is necessary and sufficient, when it is best viewed as an adjunct, and when it is irrelevant or inappropriate.
Now, to return to Bell’s key points. Firstly, she is concerned about “the stratospheric increase in antidepressant prescribing”. This disquiet about increased prescribing of a drug, echoed by many journalists, is rarely expressed in the non-psychiatric domain of medicine – and often has stigmatising undertones. If not a cost issue, the salient parameters should be utilitarian ones at the community level (is the expansion in prescribing associated with more people being helped out of their mood state?) and cost-benefit ones at the individual level (if trialled on an antidepressant, does the individual – and their clinician – judge the benefits to outweigh the costs?).
My concern is with Bell’s ad hominem “pull up your socks” stance towards many with depressive conditions. The title The Worried Well joins with a term that I have never previously heard – “misery-chic” – as misanthropic, trivialising and belittling. The risk to her assertions is that, in seeking to damn those engaging in cosmetic psychopharmacology, she unfairly demeans those with significant mood disorders. Such people already have enough of a struggle dealing with the black fog of their dis-ease, without having to doubt whether they now should continue with the medication that they had thought to be both valid and helpful.
Just as with the proverbial elephant in the room, the elephantine abstract concept of “major depression” tempts all of us to tunnel-vision – a monolithic understanding of depression and a for-or-against approach to antidepressant medication. In the same way that one might discuss footballers solely in terms of their athletic prowess, ignoring any antisocial behaviour, or vice versa, such one-eyed and stereotypical generalisations risk a lack of perspective. For some depression sufferers, antidepressant drugs are life-savers, for others life-enablers, for others of no use, and, for a percentage, a form of medication that makes things worse because of its irrelevance or side effects. The balance can be influenced by clinical sophistication in melding the art and the science. Clinicians who practise according to a sophisticated and richer model, who understand when to prescribe an antidepressant, who inform patients about salient side effects and provide pre-emptive strategies to reduce their chance or their effects, will continue to advance the management of depression. But, even in such optimal circumstances, our ability to successfully predict individual responses in advance remains unsatisfactory – but the same caveat applies to non-drug approaches to depression and to many non-psychiatric conditions.
Bell’s polemic embodies the frustrations of the disappointed and the sceptical, and perhaps her own disappointment when an SSRI failed to meet her expectations. However, extrapolation of her inner world to a world view is of greater concern when it is wedded with an austere solution. Her essay finishes with a thought that is edgy at best but puritanical in its import. Bell ponders how realistic it is to turn back the clock to when treatments were simpler, non-drug based and “traditional” (“a good night’s sleep and a bowel movement”). Medicine has allowed many choices (e.g. childbirth with anaesthesia vs natural childbirth) and we benefit from such advances. Extrapolating from romantic and simplistic visions of “the good old days” – which never were – risks shaming many into suffering in silence.
So, while I share many of Bell’s concerns, she diminishes the fact that the depressive disorders are painful and disabling, and so distorts the meaning and value of her analysis. A lack of balance, chemical or otherwise, is, at the end of the day, a lack of balance.
Gordon Parker is Scientia Professor of Psychiatry at the University of New South Wales and executive director of the Black Dog Institute, Sydney.
This correspondence featured in Quarterly Essay 19, Relaxed & Comfortable.
ALSO FROM QUARTERLY ESSAY