Bitter pills
Antipsychotics and antidepressants were once heralded as the most successful way of treating people with mental disorders. Now, a quarter of a century after their use became widespread in developed countries, doubts hang over their long-term efficacy
Over the past 30 years, the number of people using psychiatric drugs in most developed countries has increased greatly. We believe medication to be effective in treating schizophrenia, depression and bipolar disorder and believe that it 'fixes' chemical imbalances in the brain. We consider psychiatric drugs to be a success story, and yet here is the puzzle: as countries have embraced their use, the burden of mental illness has markedly grown.
In the US, for example, 1.25 million Americans aged 18 to 65 were receiving governmental payments due to mental disability in 1987. Over the next 20 years, annual spending on psychiatric drugs rose from $800m to at least $35bn. More and more people got treated, and yet the number of disabled mentally ill soared to four million.
Other countries have seen similar jumps. In Iceland, the number of new cases of adult disability due to mental disorders rose from 83 per 100,000 people in 1990 to 217 per 100,000 people in 2005, while in Britain, days of incapacity due to mental problems rose from 54 million in 1985 to 154 million in 1995. The World Health Organisation (WHO) estimates that mental illness will account for 15 percent of the global disease burden in 2020, up from 9.7 percent in 1999.
These statistics raise an obvious question: is it possible that drug-based therapy is fuelling the rise in mental illness? Psychiatry can point to evidence that its drugs are effective. In short-term trials, medications are generally more effective in suppressing the symptoms of mental disorders than placebos. The case for keeping people on medications comes from studies that show a higher occurrence of relapse in patients who come off medication than in those who stay on it. This is seen as evidence that drugs prevent the return of the 'disease'.
Flaws in the common wisdom
However, there is a hole in this evidence base. The relapse studies reveal that, once patients start taking a psychiatric drug, they are at a high risk of relapse when the drug is withdrawn. That is particularly true if the drug is stopped abruptly, which is what happens in most of these studies. But the trials do not say anything about the long-term course of patients who have never been medicated for their disorder. Nor do they provide any information about how well medicated patients are functioning over the long term in comparison with an unmedicated group.
As for the chemical imbalance theory of mental disorders, this arose in the 1960s after scientists discovered how antipsychotics and antidepressants act on the brain. They found that antipsychotics such as chlorpromazine block dopamine receptors in the brain; this led them to hypothesise that schizophrenia was caused by too much dopamine activity. First-generation antidepressants increased levels of serotonin and norepinephrine (two neurotransmitters) by blocking their removal from the synaptic cleft; this seemed to suggest that depression was caused by a deficit of these chemicals. However, scientists never found evidence of overactive dopamine systems in schizophrenia patients, nor of low serotonin levels in people with depression. "We have hunted for big, simple neurochemical explanations for psychiatric disorders and have not found them," wrote Kenneth Kendler, co-editor-in-chief of Psychological Medicine, in 2005.
In fact, these medications could be said to cause chemical imbalances. Psychotropic drugs, explained neuroscientist Steve Hyman in 1996, when he was director of the National Institute of Mental Health (NIMH), "perturb" a neurotransmitter system, and the brain then "adapts" to their presence. When people are put on an antipsychotic that blocks dopamine receptors, their brains compensate in two significant ways: the presynaptic neurons release more dopamine and the postsynaptic neurons increase the density of their receptors for that chemical messenger. When people take an antidepressant that increases serotonin activity, the opposite happens: the presynaptic neurons decrease their firing rates and the postsynaptic neurons decrease the density of their receptors for serotonin. In each case, the brain is trying to nullify the effects of the drug, a process known as 'oppositional tolerance'. At the end of this adaptive process, Hyman concluded, the brain is functioning in a manner that is "qualitatively, as well as quantitatively, different from the normal state".
To investigate the long-term effect of psychiatric medications, it is necessary to flesh out the spectrum of outcomes reported for a major mental disorder prior to the introduction of drugs. This will show whether there is a coherent narrative of science to be discovered about how the drugs affect people's lives.
With schizophrenia, the big surprise is the reported outcomes for first-episode patients hospitalised from 1945 to 1955. According to the common wisdom, chlorpromazine made
it possible to treat schizophrenia patients in the community. In fact, in the decade prior to chlorpromazine's introduction, two-thirds of first-episode patients were discharged within
18 months, 70 percent or so were living independently in the community five years after initial hospitalisation and more than 50 percent were working. Only 30 percent of the initial cohort needed continual hospitalisation.
The second surprise is found in the first longer-term study of antipsychotics conducted by the NIMH. After a year, those treated with drugs in the hospital were more likely to have been rehospitalised. This finding hinted at a paradox: medications that are effective in the short term might increase the chronicity of the disorder in the long term.
In the 1970s, the NIMH funded two studies that involved initially treating newly psychotic patients without antipsychotics, and then using drugs only if patients did not improve. Each time, using medications selectively produced better overall outcomes; each time, those who never took medication had the best long-term outcomes. These results suggested that antipsychotics were making schizophrenia patients more biologically vulnerable to psychosis than they would otherwise be. Guy Chouinard and Barry Jones from McGill University found that the drugs caused the brain to become supersensitive to dopamine. This was why drug-withdrawn patients relapsed at such high rates, and also why long-term use of antipsychotics could lead to a deepening psychosis, characterised by new and more severe symptoms.
Since then, psychiatry has done its best to avoid thinking about this problem, yet the evidence keeps piling up. In 1992, the WHO reported that schizophrenia outcomes in three developing countries, in which only 16 percent of patients were regularly maintained on antipsychotics, were much better than in seven developed countries, in which lifelong use was the standard of care. More recently, MRI studies revealed that antipsychotics shrink the brain and that this decline in brain volume is associated with emotional and cognitive decline.
Finally, in 2007, psychologist Martin Harrow published the 15-year outcomes of a large group of schizophrenia patients he had been following since their initial diagnosis. Forty per cent of those who had stopped taking antipsychotics recovered, versus 5 percent of those on medication. The unmedicated patients were much less likely to be still experiencing psychotic symptoms, and more than 50 percent were working. "I conclude that patients with schizophrenia not on antipsychotic medication for a long period of time have significantly better global functioning than those on antipsychotics," Harrow said at a meeting of the American Psychiatric Association.
The narrative for antidepressants is similar. Prior to the antidepressant era, depression was considered an episodic illness, and the majority of hospitalised patients enjoyed fairly good long-term outcomes. However, shortly after the drugs were introduced, a handful of psychiatrists observed that while their patients were now getter better sooner, they were relapsing more frequently. The drugs were causing a 'chronification' of the disorder, several psychiatrists wrote. In the early 1970s, Dutch physician JD Van Scheyen conducted a five-year study to test this possibility and concluded that long-term medication "exerts a paradoxical effect on the recurrent nature of the vital depression", increasing the frequency of relapses.
Today, depression runs a much worse course than it used to. Only 15 percent of patients treated with an antidepressant get well and stay well; the remaining 85 percent suffer from recurring episodes. Forty per cent of patients today become 'treatment resistant'. As is the case with antipsychotics, the worry is that, because of the oppositional tolerance process, antidepressants make the brain more biologically vulnerable to depression and thus, in the words of Giovanni Fava, editor of Psychotherapy and Psychosomatics, "propel the illness to a more malignant and treatment-unresponsive course".
In modern longitudinal studies, those who eschew antidepressants are regularly found to suffer from fewer periods of depression and to have lower disability rates. In a large Canadian study of workers who received short-term disability payment due to depression, 19 percent of those who took an antidepressant went on to receive long-term disability payments, versus 9 percent of the unmedicated group. Similarly, in a six-year NIMH study of 547 people who suffered a bout of depression, those who were treated were three times more likely than the untreated group to suffer a "cessation" of their "principal social role" and nearly seven times more likely to become "incapacitated".
The bipolar room
Even before we look at bipolar outcomes, we have to figure out where all the bipolar patients are coming from. Fifty years ago, bipolar illness affected one in every 5,000 adults. Today, it is said to affect one in 50 adults. Although there are many reasons for this startling increase, including the expansion of diagnostic boundaries, antidepressant-induced mania is a primary cause. In a study of 87,000 patients diagnosed with depression or anxiety, Yale Medical School investigators found that those treated with antidepressants converted to bipolar at the rate of 7.7 percent per year, three times greater than for those not exposed to the drugs. As a result, at least in the US, more than 20 percent of all patients initially diagnosed with unipolar depression eventually convert to a bipolar diagnosis.
Bipolar outcomes have notably deteriorated in the past 40 years. Like depression, in the pre-drug era it was seen as an episodic illness with good long-term outcomes. Today, patients have more episodes; they are more likely to be rapid cyclers (experiencing more than four episodes in a 12-month period); they suffer from more low-level depression between acute episodes; they are more functionally impaired and they show signs of cognitive impairment over time, which was not always the case. "Prognosis for bipolar disorder was once considered relatively favourable, but contemporary findings suggest that disability and poor outcomes are prevalent," concluded Harvard Medical School psychiatrist Ross Baldessarini in a 2007 paper.
Rethinking psychiatric medications
Psychiatric medications have an obvious utility. They can ameliorate distressing symptoms over the short term, and some people fare well on them in the long term. But the manner in which they are prescribed, as a first-line therapy for so many diagnoses, with little attention being paid to their long-term effects, has been a recipe for a medical disaster.
We need to rethink our use of these drugs, and we can start by rediscovering that psychiatric disorders can be episodic. Once that is understood, we can develop ways to help people recover from acute periods of distress, without turning them into lifelong mental patients. Next, we need to be honest about what the drugs do. They do not fix chemical imbalances,
but rather cause the brain to function in an abnormal manner. We also need to understand that the studies that are currently used to assess the drugs' merits - short-term clinical trials and relapse studies - do not provide information about their long-term effects, and that we need to shift our societal attention to that outcome.
It is easy to find alternative models of care that produce better outcomes. More than 20 years ago, western Lapland developed 'open-dialogue' therapy, which involves treating first-episode psychotic patients with intensive psychosocial support and selective use of antipsychotics, and today its patients enjoy the best outcomes in the western world. Five years after initial diagnosis, two-thirds of patients have never been exposed to antipsychotics, and 80 percent are working or back in education. As for better ways to treat depression, doctors in the UK can write a prescription for exercise, which provides the patient with access to a gym and to a personal trainer who develops an activity plan; there is good evidence that regular exercise will lead to a better long-term outcome than will taking a pill.
We would do well to rediscover our Shakespeare, too. Psychiatry and the pharmaceutical companies have worked together to define normalcy in the most restrictive way imaginable, but Shakespeare reminds us that to be emotional - and, at times, even a bit mad - is to be human. With that reference in mind, we would become more resistant to seeing ourselves as 'mentally ill'. This would put a brake on the use of drugs that have not been shown to improve long-term outcomes and that may, in fact, cause a great deal of harm.
Robert Whitaker is a journalist and author.
Artwork: Gregory Iliopoulos
Co-production in mental health services "Our work aims to engender social inclusion by involving service users and the public in the design and delivery of recovery services," said Morris. "There has been comparatively little focus on this approach before now." |
Share