Wednesday, February 23, 2011

Gold Standard Clinical Trials - Science or Marketing?: Part IV

This is the fifth in a series that takes a hard look at the information that we and our clinicians rely on to give us an insight into our condition and how to treat it. In the first piece, we reviewed three unscientific reader surveys I conducted here at Knowledge is Necessity, rightly at the very bottom of the evidence food chain but that actually told us something. The next three articles (Part I, Part II, Part III) looked at a “randomized, double blind, placebo-controlled” (RDPC) clinical trial that tested J&J's Invega for schizoaffective - regarded as the gold standard of research - but which actually told us nothing.

Moving on ...

In psychiatry, RDPCs imitate science rather than embody science, and therefore need to be regarded with a very high degree of skepticism. Two principles are in play:
  1. We know next to nothing about the phenomenon we are investigating.
  2. RDPCs are conducted for reasons of marketing.
We Know Next to Nothing

What is depression? Feeling sad? Loss of pleasure? Unmotivated? Psychic numbness? Reaction to stress? Loss of energy? Over-ruminating? Under-thinking? On and on, it goes.

In all likelihood, we are talking many - possibly hundreds - of illnesses involving by far the most complex organ in the body and all its related systems, with myriad possible causes and effects (including genetic vulnerability to environmental stressors) recruiting a host of different pathways. RDPCs do not recognize this. Rather they test “antidepressants” on a “depressed” population.

The equivalent would be to treat all infectious diseases as the same, call every such outbreak “runny nose disease,” and run a clinical trial of the same test med on everyone with a runny nose or high fever. Any reasonable bystander would rightly regard this as insane. Why, then, would anyone apply this approach to testing psychiatric meds? Read on ...


Only pharmaceutical companies can afford the $40 million or so (my best guess) it takes to conduct the two clinical trials required for an FDA indication to treat an illness, but the pay-off is huge. For a common chronic illness, such as “depression,” an FDA indication amounts to a license to print money.

The drug companies are abetted by the FDA, which sets a very low success threshold for a clinical trial. In the case of antidepressants, assuming safety and tolerability standards are met, a mere 50 percent of those in the drug group need to get merely 50 percent better. (For antipsychotics to test schizophrenia, the standard is even lower.) You can achieve nearly as good a result with a placebo - literally. But why would a drug company care about that? Once they have succeeded in putting a tiny bit of daylight between their drug and the placebo in two trials, they now have their license to print money.

What RDPCs Do Tell Us

Of all things, despite the best efforts by drug companies to produce results favorable to their test med, the one thing clinical trials involving antidepressants do show us is their spectacular lack of effect on wide populations classified as “depressed.” Commentators such as Robert Whitaker have had a field day with this. Likewise Whitaker et al can point to the indisputable fact that these meds have failed to make a dent in the worldwide “depression” epidemic.

There is no way to “refute” or “repudiate” this evidence, except with equally weighty evidence, which does not exist. We can make a case that antidepressants are likely to work extremely well on a certain small subpopulation of patients, but to prove that we would first need to identify this population and then test them in a randomized, double-blind, placebo-controlled clinical trial.

Ironic, isn’t it?


Gledwood said...

I have suffered depression since my late teens and possibly since childhood. I do remember clearly at least 2 of the popular psychiatry/self-help books of the time listing Depression TWICE. Under Neurotic or Reactive Depression and Psychotic or Endogenous Depression. While the books acknowledged that many if not most depressions entailed aspects of both types it was very strange having to decide between two differently slanted descriptions of your own problem and find out which was most likely "me"...

John McManamy said...

Hey, Gledwood. This is very interesting. These books would have come out when the DSM-II of 1968 was in force. You can make a very good case that the DSM-II view of depression is far more nuanced. The modern DSM went for an over-simplified view. I would argue for some kind of reconciliation with brain science etc folded in. My point is the DSM approach encourages both patients and doctors to stop thinking once we are dx'd with depression. I say the dx is where the thinking needs to start.