Curing depression

The Guardian recently reported a “ground-breaking” study by a team led by Professor Andrea Cipriani showing that “antidepressants work.”   The debate is over, said The Guardian, and there is no conspiracy.  Antidepressants work, and we should rejoice.

The real story is more nuanced than The Guardian writers would have us think.  To begin with, the study broke no new ground at all.  It was a meta-analysis of existing research, which by definition breaks no new ground.  Further, the study found that the very best antidepressant, amitryptiline, is slightly more than twice as likely to work than a placebo, and the rest work only marginally better than placebo.  Detailed critiques of the Cipriana study have been published by academic psychiatrists such as Joanna Moncrieff, and journalists such as Robert Whitaker.  What the ground-breaking study really demonstrated is that antidepressants, at best, only barely work. 

This bare improvement over placebo comes at a huge cost.  Antidepressants come with side effects ranging from weight gain and hair loss through to suicide and various kinds of persistent or permanent disablement.  Long term use of amitryptiline, for example, comes with a 50% increase in the risk of dementia.  Without exception, every person who takes an antidepressant will experience side effects.  Placebos, conversely, are nearly always entirely free of side effects.

If every person who has been given an antidepressant over the last few decades had instead been given a placebo, almost as many people would have gotten relief from their depression, but with barely any side effects.   Mindfulness meditation has been shown repeatedly to work at least as well as antidepressants.  If every doctor in the world immediately stopped prescribing antidepressants for all but the most severe cases of depression, and instead prescribed mindfulness meditation and placebo, vastly more patients would be better off and there would be no disease burden from side effects.   

But the story doesn’t end there.  One of the side effects of antidepressants is mania.  This term refers to what happens when a person completely loses control of their mind and behaviour.  A person who is suffering from mania frequently ends up in hospital.  There, the doctor observes that the person was depressed, and is now manic.  According to the Diagnostic and Statistical Manual of Mental Disorders (6^th edition), that person can be officially diagnosed as having bipolar disorder despite the fact that the mania was caused by an antidepressant. 

A person suffering depression is not normally diagnosed with a lifelong condition.  But a person who is diagnosed with bipolar disorder is told that they have the condition for life, and is then prescribed any number of expensive and disabling medications.  Their lives are shortened by an average of 20 years, and many will exit the workforce as disability pensioners.  Instead of having a customer for a few months, the pharmaceutical industry now has a customer for life, with the costs of expensive medications borne mostly by the taxpayer.   

How often does an antidepressant cause mania?  Researchers have offered figures of between 3% and 15%, depending on the antidepressant.  For argument’s sake, let’s say that the average percentage is 5%, and 1 in 20 people who take an anti-depressant will become manic as a direct result.  If doctors told their patients that there was a 1 in 20 chance that they could become manic from taking a drug, and that on becoming manic they could be hospitalised and diagnosed as bipolar, how many people would take the drugs? 

Peter Breggin, a retired American psychiatrist, has written that when he commenced medical practice many decades ago a diagnosis of bipolar disorder was very rare.  Now it is one of the most common diagnoses, and Dr Breggin, like many other qualified writers on the subject, blames antidepressants. 

It is an unfortunate fact that doctors in Australia are not required to report their diagnoses of mental illnesses.  We know how many prescriptions for antidepressants are written each year, but we do not know for certain how many people are diagnosed with bipolar disorder each year.  Since 2008, however, we have kept statistics on hospital admissions.  If antidepressants are causing diagnoses of bipolar disorder, the rate of increase in admissions for bipolar disorder should be similar to the rate of increase in prescriptions for antidepressants.

The Australian Institute of Health and Welfare publishes data on both mental health admissions and antidepressant prescriptions.  For the period between 2009/10 and 2014/15, data is available for both, and the increases can be compared.  During that period, antidepressant prescriptions increased by 17%, and bipolar disorder diagnoses in hospitals by 12%.  The number of schizophrenia diagnoses, by contrast, was almost unchanged over the same period. 

Correlation does not imply causation, and a bare comparison such as this is not a rigorous statistical analysis.  Nonetheless, from these statistics, a prima facie case can be made that the increase in antidepressant prescriptions is causing an increase in bipolar disorder diagnoses.  We know that mania is a side effect of antidepressants – this is admitted by the pharmaceutical companies, established beyond doubt by the research, and is not a matter for dispute.  

The official pharmaceutical industry position on antidepressant-induced mania is that antidepressants do not cause mania, but rather they cause an underlying bipolar condition to manifest itself.  It is not possible to falsify such a proposition, so the official position is unscientific, but it is repeated constantly, accepted as true, and it enables the industry to continue aggressively marketing antidepressants.

According to the precautionary principle (which states that when an activity raises threats of harm to human health, precautionary measures should be taken even if a cause and effect relationship is not fully established) we would be justified in restricting the use of antidepressants immediately.  At the very least we should be looking very closely at whether the correlation is in fact a causal one. 

Nearly all antidepressants are prescribed by GPs.  In order to establish whether or not antidepressants are causing the increase in bipolar disorder diagnoses GPs should be required to conduct mandatory follow-up consultations and report on the side effects of antidepressants as a condition of prescribing them.  A mandatory post-prescription reporting regime would take up very little doctors’ time but would enable us to determine for certain whether or not antidepressants are responsible for the modern epidemic of bipolar disorder.   Alternatively, a publicly-funded outcomes study would settle the matter.