Easy to study short term effects, but the long term is what matters

A week ago or so I posted a review of Robert Whitaker’s book Anatomy of an Epidemic on Amazon.com.  The purpose of doing so was to address what some have seen as a problem of Whitaker relying too much on “naturalistic” studies in coming to his conclusion that psychiatric medications are doing more harm than good on average.

What I said in response to that contention was that 

I think not.  What Whitaker perhaps assumes everyone knows, but doesn’t discuss, is the fact that it is really impossible to perform tightly designed scientific studies about long term outcomes.  People will not stay in a double blind randomized study over the course of years, so such studies just don’t and can’t happen.  Because of this, we are left in a bind, with two choices:

• 1.  We can base our decisions on the more “scientific” studies, that have tight designs but look only at short term outcomes, usually just weeks or months, or
• 2.  we can base our decisions more on naturalistic studies, which are somewhat less scientifically convincing but at least are studies of what we really care about, the actual long term outcomes, what happens over years or even decades.

If we choose option 1, we fall victim to something called the “Streetlight Effect” which is described by Louis F. Markert as follows:  “The Streetlight Effect is the fundamental error of searching for something where the light is best, but not where it actually may lie. It is described in a favorite old joke among scientists–a fact which alone should raise our eyebrows! Late one night a policeman finds a drunk crawling around on his hands and knees under a streetlight. The drunk tells the officer he’s looking for his wallet. When the officer asks if he’s sure this is where he dropped his wallet, the man replies that he thinks he actually dropped it across the street. “Then why are you looking over here?” the officer asks. “Because the light’s better here,” explains the drunk!”

When a person has a psychotic episode for the first time, or when a person is already taking antipsychotics and is wondering whether or not to continue, the most important consideration is most likely to contribute to a positive long term outcome for the person.  But we have no way to study that with randomized double blind studies.  The streetlight of randomized double blind studies does not shine on the long term.  It shines only on the short term, and shows us that in that short term, people are better off getting on or staying on the drugs.  As far as psychiatry has been concerned, that should be the end of the story, and everyone with psychosis should stay on the medications for life.

What Whitaker has done has been to systematically look at the studies that go beyond the short term, to see what they indicate.  These studies are a bit murkier, like views away from the streetlight, and their interpretation is more problematic.  In any one study, it is always possible that some factor or other contaminated the result, making the study appear to show one thing when really something else was happening.  But these are the studies that at least actually look at the outcome we really care about, so they are the best evidence we have to determine what is happening.  And when they all say essentially the same thing – that long term outcomes for people who take the medications are worse on average than for those who do not take them – then that is strong evidence we need a radical change in direction in mental health treatment.

It would be much simpler and easy for all of us if what reduced psychosis most in the short term was also what best supported long term recovery and mental and physical health.  But it’s time for the entire mental health system to take note that this doesn’t seem to be the case, and so we need to be looking for ways to help those who want to attempt safe tranistions to less or no medications, and we need to work on really understanding people rather than assuming we can forever use medications to “make people normal.”