Which came first, the condition or the drug?

Early in the morning on 13 December 2006, police officers from the small town of Hull, Massachusetts, near Boston, arrived at the house of Michael and Carolyn Riley in response to an emergency call. Their four-year-old daughter, Rebecca, had been diagnosed with bipolar disorder two years earlier. When the officers reached the house, they found Rebecca sprawled on the floor next to her teddy bear. She had died from an overdose of the medication cocktail prescribed by her psychiatrist, Kayoko Kifuji. At the time of her death, she was taking Seroquel, a powerful antipsychotic drug, Depakote, a no less powerful anticonvulsant and mood-stabilising drug, and Clonidine, a hypotensive drug used as a sedative.

Rebecca’s parents were charged with first-degree murder, but her doctor’s role must also be questioned. How could she have prescribed a two-year-old psychotropic medications normally intended for adults suffering from psychotic mania? Yet the medical centre where Rebecca had been treated issued a statement describing Dr Kifuji’s treatment as ‘appropriate and within responsible professional standards’. In an interview with the Boston Globe, Janet Wozniak, director of the paediatric bipolar programme at Massachusetts General Hospital, went even further: ‘We support early diagnosis and treatment because the symptoms of [bipolar] disorder are extremely debilitating and impairing … It’s incumbent on us as a field to understand more which pre-schoolers need to be identified and treated in an aggressive way.’ On 1 July 2009, a Plymouth County grand jury dropped all criminal charges against Dr Kifuji. Carolyn and Michael Riley were respectively convicted of second and first-degree murder earlier this year.

How did we come to this? In Mania: A Short History of Bipolar Disorder, David Healy goes back to the Greeks and Romans in search of an explanation. There are good reasons for this. Very few people had heard of bipolar disorder before 1980, when it was introduced in the DSM-III – the diagnostic manual of the American Psychiatric Association – and it was only in 1996 that a group of doctors from Massachusetts General Hospital, led by Joseph Biederman and Janet Wozniak, first proposed that some children diagnosed with attention-deficit/hyperactivity disorder might in fact suffer from bipolar disorder. But whoever googles ‘bipolar disorder’ today is likely to learn that the illness has always been with us. It’s just a new name, we are told, for what used to be called manic depression, a mood disorder characterised by oscillations between states of manic hyperactivity and deep depression which had already been described by Hippocrates and other ancient physicians. It is often added that Newton, Van Gogh, Schumann and Boltzmann suffered from the disorder, and even that they owed their genius to it.

Healy has no trouble demonstrating that this history is a retrospective illusion. While the Greeks did indeed speak of mania and melancholia, these terms covered all sorts of hyperactive deliriums and lethargic stupors, the majority of which were probably caused by infectious or post-infectious states, or perhaps by Parkinson’s disease or hypothyroidism. Ancient physicians limited themselves to the visible physical manifestations of the diseases they described, so we have no way of knowing if these corresponded to what we today categorise as bipolar disorder. Even when they established a link between melancholia and mania without fever, as Soranus of Ephesus and Aretaeus of Cappadocia did, they may well have been describing fluctuations of psychotic agitation unrelated to mood swings.

Bipolarity in the modern sense could not have emerged until it became possible to identify mood disorders without delirium or intellectual disorders; in other words, it required a profound redefinition of what had until then been understood as madness or insanity. This development started at the beginning of the 19th century with Esquirol’s ‘affective monomanias’ (notably ‘lypemania’, the first elaboration of what was to become our modern depression) and led in 1882 to Kahlbaum’s ‘cyclothymia’ and ‘dysthymia’, two mood disorders that he firmly distinguished from what he called ‘cyclical insanity’ (Irresein). The identification of bipolarity also required the researcher to be in a position to observe and compare the course of various psychical disorders over the long run, which, as Healy points out, could be done only once mentally ill people were gathered in asylums. Before this, it would have been impossible to identify, as Jean-Pierre Falret and Jules Baillarger did independently in 1854, a ‘circular insanity’ or folie à doubleforme characterised by a regular alternation of manic excitation and melancholic depression. In 1899, Emil Kraepelin wove all these threads into what he proposed to call ‘manic-depressive insanity’, to distinguish it from dementia praecox (later renamed ‘schizophrenia’ by Eugen Bleuler).

Healy, though, does not describe Kraepelin’s manic-depressive insanity as the culmination of a continuous evolution. On the contrary, he shows that the nosographic systems of Falret, Kahlbaum, Kraepelin and others were in fact different and incommensurable, each one drawing up the map of signs and symptoms according to different criteria. The x they were trying to define was a moving target – and it has remained so. One thing, however, seems clear: whatever this illness was, it was relatively rare. On the basis of 3872 admission charts from the asylum at Denbigh, North Wales, between the years 1875 and 1924, Healy arrives at a figure of ten cases per million each year, that is 0.001 per cent of the general population. This figure is striking, as today the incidence of bipolar disorder is supposed to be much higher. In 1994, for example, the US National Comorbidity Survey estimated that 1.3 per cent of the American population suffered from bipolar disorder. Four years later, the psychiatrist Jules Angst upped the figure to 5 per cent: 5000 times higher than the figure suggested by Healy. And yet we continue to be told that bipolar disorder is merely a new name for the old manic depression. Are we really talking about the same thing (the realist hypothesis)? Or did the name create a new thing (the nominalist hypothesis)?

Healy favours the nominalist hypothesis (which doesn’t mean that he denies the ultimate reality of x; quite the contrary). The term ‘bipolar disorder’, he explains, was introduced in 1966 by Angst and by Carlo Perris. Both were influenced by the work of Karl Kleist and Karl Leonhard on the distinction between ‘unipolar’ disorders, which always present themselves in the same way, and ‘multipolar’ disorders, which manifest themselves differently in each case. Angst and Perris both proposed cleanly separating unipolar depressions from bipolar disorders (they were contradicting Kraepelin, who believed that both sets of disorders were presentations of one and the same manic-depressive illness). While their conceptual move has been hailed as a breakthrough, it is hard to understand what the point is: it seems to muddle the diagnosis instead of clarifying it. In practice, how are we to distinguish a unipolar depression from a bipolar disorder in a patient who has yet to experience a manic episode? Nonetheless, instead of seeing this incoherence as a reason for rejecting the new paradigm, psychiatrists have done their utmost to patch it up with all sorts of ad hoc nosographic innovations.

First, a distinction was made between ‘bipolar I’, which applied to patients hospitalised for both depressive and manic episodes, and a brand new ‘bipolar II’, which referred to patients hospitalised solely for a depressive episode. In other words, any person hospitalised for depression could now be diagnosed as bipolar. Then the reference to hospitalisation was dropped from the description of bipolar disorder II, which meant it could now include less severe forms of depression and hyperactivity, as well as all sorts of neurotic disorders that Kraepelin would never have dreamed of calling manic-depressive insanity. One now speaks of a ‘bipolar spectrum’, which includes, along with bipolar disorders I and II, cyclothymia (a mild form of bipolar II) and bipolar disorder ‘not otherwise specified’ (an all-purpose category in which practically any affective instability can be placed). The spectrum also includes bipolar disorders II1/2, III, III1/2, IV, V, VI, and even a very accommodating ‘subthreshold bipolar disorder’.

The category has expanded so much that it would be difficult to find anyone who couldn’t be described as ‘bipolar’, especially now that the diagnosis is liberally applied to people of all ages. Conventional wisdom once had it that manic depression burns out with age, but geriatric bipolar disorder is now the talk of psychiatric congresses. Elderly people who are depressed or agitated find themselves diagnosed with bipolar disorder for the first time in their lives and are prescribed antipsychotics or anticonvulsants that have the potential to drastically shorten their life expectancy: according to David Graham, an expert from the Food and Drug Administration, these psycho-tropic medications are responsible for the deaths of some 15,000 elderly people each year in the United States. Since the work of Biederman and Wozniak it has also been thought that bipolar disorder can strike in early childhood and not just with the onset of adolescence, as psychiatric manuals previously said of manic depression. As a result, the prevalence of paediatric bipolar disorder multiplied by a factor of 40 between 1994 and 2002. In August 2002, a cover of Time magazine read: ‘Young and Bipolar – Once called manic depression, the disorder afflicted adults. Now it’s striking kids. Why?’ The article listed a series of ‘warning signs’ for parents: ‘poor handwriting’, ‘complains of being bored’, ‘is very creative’, ‘intolerant of delays’, ‘curses viciously in anger’, ‘elated or silly, giddy mood states’.

How did we come to apply such a serious diagnosis to vaguely depressed or irritable adults, to unruly children and to nursing-home residents? Is it simply that psychiatric science has progressed and now allows us to detect more easily an illness that had previously been ignored or misunderstood? Healy has another, more cynical explanation: the never-ending expansion of the category of bipolar disorder benefits large pharmaceutical companies eager to sell medications marketed with the disorder in mind. Psychiatric research doesn’t evolve in a vacuum. Behind the psychiatrists’ constant redrawing of the map of mental illness in a sincere effort at better understanding, there are enormous financial and industrial interests that steer research in one direction rather than another. For researchers, mental illnesses are realities whose contours they attempt to define; for pharmaceutical companies, they are markets that can be redefined, divided up and extended in order to make them ever more lucrative. The uncertainties of the psychiatric field present a magnificent commercial opportunity, since illnesses can always be tailored to sell a particular molecule under a particular patent. This is what industry insiders call ‘condition branding’. As Vince Parry, the president of an advertising agency and author of professional articles on pharmaceutical marketing, puts it, ‘No therapeutic category is more accepting of condition branding than the field of anxiety and depression, where illness is rarely based on measurable physical symptoms and, therefore, open to conceptual definition.’

In the case of bipolar disorder, this conceptual gerrymandering has involved stretching and diluting the definition of what used to be called manic-depressive illness so that it might include depression and other mood disorders, thus creating a market for antipsychotic or anticonvulsant medications that were initially approved only for the treatment of manic states. Healy cites an advertisement on American TV in which a young woman dances joyfully in a night club; a voiceover comments: ‘Your doctor never sees you like this.’ The next shot shows the same woman, now very glum: ‘This is who your doctor sees.’ Then we see her frantically shopping in a department store. The voiceover again: ‘That is why so many people with bipolar disorder are being treated for depression and aren’t getting any better – because depression is only half the story.’ No medication is mentioned, but the viewer is invited to consult a ‘mood disorder questionnaire’ on the website Bipolar Help Center, sponsored by Lilly Pharmaceuticals.

Lilly happens to be the maker of Zyprexa. Zyprexa, just like AstraZeneca’s Seroquel and Janssen’s Risperdal, is an antipsychotic originally approved for the treatment of schizophrenia. It was only subsequently that the companies producing these ‘atypical’ antipsychotics obtained licences to market their drugs for the treatment of bipolar disorder and, by extension, mood disorders in general. The same job was done on anticonvulsant medications, which are strong sedatives prescribed for epileptic attacks. In 1995, Abbott Laboratories succeeded in obtaining a licence to offer its anticonvulsant drug Depakote for the treatment of mania. This was hardly difficult because, as Healy points out, ‘giving any sedative to manic patients will produce a benefit on suitable rating scales’. Depakote, however, was marketed not as an anticonvulsant but as a ‘mood stabiliser’ – a term without any clinical meaning and misleading insofar as it suggests a preventive action against bipolar disorder that has never been established in any study.

In the wake of this ‘conceptual redefinition’, other anticonvulsants such as Warner-Lambert/Parke-Davis’s Neurontin were aggressively marketed for mood disorders when they hadn’t been approved even for manic states. But what did it matter, since the meteoric success of the concept of ‘mood stabilisation’ made this step useless? The suggestion to doctors was that they prescribe anticonvulsants or atypical antipsychotics to ‘stabilise’ the mood of patients who had never before displayed any manic hyperactivity. Anyone who knows how lucrative the market was for selective serotonin reuptake inhibitor (SSRI) antidepressants such as Prozac or Seroxat in the 1990s will immediately see the point of the exercise. While most SSRIs are now off patent, the market for atypical antipsychotics is currently worth $18 billion – twice as much as it was for antidepressants in 2001.

The question is whether the pharmaceutical industry’s marketers actually created bipolar disorder or merely exploited tentative psychiatric research. Strictly speaking, we must grant it was opportunism: the research of Angst and Perris on bipolar disorder dates from 1966, well before the development of atypical antipsychotics and ‘mood stabilisers’. But the reality of the contemporary medical-industrial complex is that their hypothesis would not have survived, let alone prospered, had it not been ‘recruited’ at a particular moment by the pharmaceutical industry and thrust on the public with the help of the most sophisticated marketing and advertising techniques.

This is what Healy calls the ‘manufacture of consensus’, echoing Walter Lippmann’s and Edward Bernays’s ‘manufacture of consent’. By subsidising one research programme instead of another, one conference or symposium, one journal, one publication, one learned society etc, the pharmaceutical industry makes precious allies among the ‘key opinion leaders’ of the medical establishment, and gains a very efficient means to steer academic discussion towards the illnesses that interest it at any given moment.

Healy tells the story of the launch of bipolar disorder at the end of the 1990s. A specialised journal, Bipolar Disorder, was established, along with the International Society for Bipolar Disorders and the European Bipolar Forum; conferences were inundated with papers commissioned by the industry; a swarm of publications appeared, many of them signed by important names in the psychiatric field but actually ghost-written by PR agencies. Once the medical elites were bought and sold on the new disease, armies of industry representatives descended on clinicians, to ‘educate’ them and teach them how to recognise the symptoms of bipolar disorder. Bipolar patient advocacy groups sprang up, generously supported by pharmaceutical companies; freelance journalists were hired to write magazine articles on the latest advances in psychiatric science; websites were created, such as IsItReallyDepression.com (sponsored by AstraZeneca), where you can fill out a ‘mood assessment questionnaire’ at the end of which you’ll inevitably be dispatched to the nearest doctor. As a British blogger noticed recently, the Wikipedia entries ‘Bipolar Disorder’ and ‘Bipolar Spectrum’ were edited from a computer belonging to AstraZeneca, ensuring that everyone is on the same diagnostic page as the industry.

After this marketing blitz, it was no longer possible to ignore bipolar disorder. A new reality emerged, constructed by scientific articles, patient testimonials, rating scales, statistics, epidemiological surveys, clinical guidelines, everyday language (‘I’m bipolar’), vague fears and reassuring medications. ‘It is essentially like setting a snowball rolling down a hill,’ a Practical Guide to Medical Education intended for industry marketers explains. ‘It starts with a small core of support: maybe a few abstracts presented at meetings, articles in key journals, focuses for discussion amongst “leading experts” … and by the time it reaches the bottom of the hill the noise should be coming from all sides and sources.’ Pharmaceutical companies today launch diseases in the way that fashion companies launch new brands of jeans, creating needs that align with industrial strategies and the duration of patents.

The techniques Healy describes are the same as those used by the pharmaceutical industry to sell, or oversell, conditions as diverse as depression, social phobia, metabolic syndrome, attention-deficit/hyper-activity disorder, fibromyalgia, premenstrual dysphoric disorder, panic attacks, restless legs syndrome and so forth. In each case, the existence and risks of one condition or another are amplified in order to persuade us to swallow chemical products that may be either useless or, often, potentially toxic.

In the case of bipolar disorder, the medications on offer come with significant risks. Anticonvulsants are liable to cause kidney failure, obesity, diabetes and polycystic ovary syndrome, and they are among the most teratogenic drugs. Atypical antipsychotics, once reputed to be less toxic than first-generation ‘typical’ antipsychotics, are now known to have very serious side effects: significant weight gain, diabetes, pancreatitis, stroke, heart disease and tardive dyskinesia (a condition involving incapacitating involuntary movements of the mouth, lips and tongue). They can, in some circumstances, cause neuroleptic malignant syndrome, a life-threatening neurological disorder, and akathisia, whose sufferers experience extreme internal restlessness and suicidal thoughts. ‘All available studies on the longer-term consequences of antipsychotics indicate that they probably reduce life expectancy,’ Healy concludes. Prescribing such toxic medications to patients suffering acute mania may, of course, be unavoidable. But as a prophylactic to be given to depressed pensioners and hyperactive children?

A series of prominent lawsuits has been brought over the past few years in the United States against the manufacturers of anticonvulsants and atypical antipsychotics for having hidden their side effects and for having marketed them ‘off label’ towards patient populations not approved by the FDA. The sums paid out in fines or settlements by the pharmaceutical companies involved are staggering, and they give an idea of how disastrous the effects of their medications have been: Warner-Lambert/Parke-Davis (now Pfizer) has paid more than $430 million for marketing Neurontin for bipolar disorder; Lilly had to pay a total of $2.6 billion for the illegal marketing of Zyprexa; Pfizer was forced to pay $301 million for the illegal marketing of the atypical antipsychotic Geodon. AstraZeneca has agreed to pay $520 million to settle federal investigations into its marketing of Seroquel and has already spent $593 million in legal fees defending itself against the 10,381 individual lawsuits brought by patients for the side effects caused by the drug. Johnson & Johnson and its subsidiary Janssen have been sued by nine American states for the off-label marketing of Risperdal.

The marketing of bipolar disorder itself has not been put on trial, and probably never will be. This is the perfect crime. Bipolar disorder I, II, III etc remain on the books and doctors continue to exercise their freedom of judgment in prescribing Zyprexa and Seroquel off label to their ‘bipolar’ patients. An extended release version of Seroquel, Seroquel XR, was approved in December 2009 by the FDA for the treatment of depression. As for the sales of Zyprexa, they are up 2 per cent compared to 2007, when the medication generated $4.8 billion in sales. Who remembers Rebecca Riley now?