Roger Shepard’s above illustration shows two tables of exactly the same size and shape. It’s an extraordinary example of how even when you know that the table tops are the same, the data changes nothing. The dynamics of perspective mean we continue to see things in the wrong way.
Early on in the Prozac and Suicide controversy, Eli Lilly adopted a strategy that has “put things in perspective” ever since – they asked us to blame the disease rather than the drug. This has framed how we see things since in quite astonishing ways.
While ghostwriting an article on Paxil for GlaxoSmithKline, Sally Laden, perhaps the famous ghostwriter of all, referred to some studies as providing data that no spin could overcome (despite her best efforts to do so). As a tribute to her, it seems a good idea to draw attention to the Spin that no Data can overcome.
When an engine falls off a plane or the plane falls out of the sky, we can spot something that shouldn’t be happening. When something goes wrong on a drug, it can be difficult to decide if it is the drug or the disease. If a drug for weight loss such as fenfluramine causes valvular heart disease, it’s difficult to argue that obesity causes valvular heart disease. But at the opposite extreme it is all too easy for drug companies to portray depression as giving rise to suicidality.
In between lies an area that companies have been busy colonizing ever since Lilly introduced the disease defence, and with time it’s become easier and easier for them to argue that there is some disease somewhere that has caused whatever problems might have been linked to their drug; we have been told schizophrenia causes diabetes among other things.
One way to reframe the issues would be to restore the word “poison” to our vocabulary. Every drug, as one of the fathers of medicine, Paracelsus, said, is a poison, and the art of medicine lies in finding the right dose. This is one of the most famous and hallowed definitions of medical practice but it has become close to illegal to voice it.
In expert reports on drug-induced injury cases, when I refer to all drugs being poisons, the defense lawyers file a motion to have the term poison removed as prejudicial and the judge will commonly rule in their favour. I’m waiting for the argument that it’s unconstitutional to refer to any medicine as a poison.
The bias introduced by a disease perspective unbalanced by a poison perspective is extraordinary. In the case of the antidepressants, the bias is almost complete. There are studies in which healthy volunteers have become suicidal or committed suicide on antidepressants – and the response from companies and their experts is that these volunteers must have been depressed or bipolar.
After the Black Box warning made it more difficult to blame depression, the antidepressants-causing-suicide story was reframed as an error in diagnosis; these patients in fact had a different disease – bipolar disorder – and if they had only been given a mood-stabilizer rather than an antidepressant, all would have been well.
This story has worked a treat – even though in bipolar patients, anticonvulsants given as mood-stabilizers double the rate of suicides and suicidal acts compared to placebo. It works even though people with epilepsy or migraine given these drugs have double the rate of suicidal acts on them compared to placebo. It works even though healthy volunteers become suicidal on them.
Despite this, doctors and academics cannot bring themselves to see medicines as poisons to be used with care. Instead, the disease has become the unique source of toxicity and this frames our current view of events.
The role of the disease as a source of toxicity has huge implications for randomized controlled trials, widely trumpeted as the gold standard method of evaluation.
The failure of doctors to spot drug toxicity has huge implications for the continuing existence of medicine as a profession.
We will explore these in my next blog posts.Share this:
Copyright © Data Based Medicine Americas Ltd.
[…] But what happens when both the poison and the disease produce superficially similar problems – when both an antidepressant and depression produce suicidality for instance? In this case there may be almost nothing that can be done to persuade people that it’s the poison and not the disease producing the effect. Even people of goodwill may look and look at the data and still not make the connection, as with the Shepard illusion (demonstrated in my previous post The Spin that no Data can overcome). […]
Thank you for clarifying this logic. You are truly a philosopher. It helps me to articlulate the truth.