More on Thinking Backwards

So, if we're thinking backwards from conclusions to evidence, where do we get the conclusions from? In my previous post on running the brain in reverse, I suggested two possibilities: 1) The conclusion is something we would like to be true, because it's convenient, evades the hard work of thinking through complexity, pleases equally backward-thinking constituents, enables us to bamboozle people, or whatever ground of selfishness or wishful thinking there may be; and 2) We just grab a set of beliefs out of the air, for example, "The Bible is the literal and inerrant word of God," and take it from there. The benefit of the latter process is not as obvious, since the belief system could well impose costs and certainly won't avoid hard thinking. The Bible, after all, is riddled with self-contradiction and is manifestly inconsistent with observable reality, so theologians have to do a lot of hard work of obfuscation and sophistry. (For a good time, check out The Skeptic's Annotated Bible.)

But today I want to talk about a third way of generating conclusions as a basis for backward thinking, and that is the reification of hypotheses.(And guess what, I'm back now on the psychiatry thread.) As we've mentioned, you will see advertisements on television explaining that the disease called depression is caused by a chemical imbalance in the brain, which antidepressants correct. Antipsychotic drugs are not advertised on television, as far as I know, but their makers also promote the idea that they correct a different fault in the neurotransmitter system.

You can read a passionate, radical attack on these ideas here, by Robert Whitaker, a journalist and author of Mad in America, and a somewhat more restrained and scholarly dissent here, by a psychiatrist and clinical social worker.

In a pistachio shell, both classes of drugs originated when researchers noticed that people taking antihistamines in clinical trials reported subjective side effects such as drowsiness and agitation. A lot has been invested in subsequent decades to develop antihistamines that don't have these effects, but meanwhile drug companies decided to find out if they could take advantage of them for use in psychiatric disease. The first drugs to result were antidepressants that interfere with enzymes in the brain that break down certain neurotransmitters; and antipsychotic drugs that block a class of receptors for the neurotransmitter dopamine.

These were considered useful because, in the case of antidepressants, slightly more people on the drug than on placebo reported improvement in the short term in formal "rating scales" for depression. These are a series of questions or observations which, when summed up, are supposed to reveal the severity of depression. Here is the most commonly used, the Hamilton Rating Scale for Depression. Note that denying being depressed is considered a symptom of depression. You get 2 points for denying being depressed, but 0 points if you say you are depressed and ill. That's interesting, I'm already part way there, I guess. You also get points for insomnia, lack of appetite, anxiety, agitation, etc. Scroll down to the bottom and notice who paid to print this form and post it on the web.

In the case of antipsychotics, the drugs suppressed the so-called active symptoms -- hallucinations, delusions, disorganized thinking and bizarre behavior. Existing treatments already did this -- the first drug used by psychiatrists for psychosis was morphine, and other sedatives were effective in calming people down. Psychiatrists also used brain surgery -- severing the frontal lobes so that people became tractable zombies -- and electric shocks to the brain or insulin-induced comas which left people groggy and passive. The antipsychotics were considered an advance since they weren't as obviously addictive as opiates and were less violent than other methods.

Noticing that these drugs had effects on neurotransmitter systems led to the hypothesis that the diseases of depression and schizophrenia were caused by flaws in these systems that the drugs were correcting. Hence further drug development research has focused on finding compounds that affect the same systems in more specific ways, or by other mechanisms. Hence the "Selective Serotonin Reuptake Inhibitors," which concentrate their effect on a single neurotransmitter; and the "atypical antipsychotics" which blockade dopamine receptors by a different mechanism.

But it turns out that if you bother to look at the actual basic research, there is no evidence that people diagnosed with these diseases have abnormalities in these neurotransmitter systems. In fact, the drugs cause abnormalities. In the case of depression, any compound which has some sedative effect will result in improvement on the Hamilton Rating Scale. And there is no particular reason to think that the effect on psychotic symptoms of neuroleptic drugs is somehow more fundamental than the effects of earlier treatments. If you don't like what the brain is doing, and you broadly suppress some of its activities, you might find the result an improvement, but that doesn't mean either that you now understand the root of the problem, or that you are treating it.

And therein lies the real offense. It would be one thing if antidepressants actually helped people, but there is affirmative evidence that they are in fact counterproductive. The few trials that have been done comparing antidepressant treatment with counseling only, find that people who are treated with antidepressants have a worse long-term course than people who are not. They relapse more often and are more depressed. These conclusions aren't quite rock solid because the studies are retrospective. Nobody wants to fund a randomized controlled trial that might prove this. But it is notorious that in some RCTs, antidepressants have actually done worse than placebo even in the short term. These studies are ignored or explained away.

In the case of antipsychotic medications the story is less clear cut. Some people do have psychotic episodes from which they recover spontaneously; and schizophrenia also tends to remit somewhat in middle age. So it's not necessarily a terrible doom if untreated, although for many people, alas, it is. The question is whether the long-term course of schizophrenia is worse for people who take antipsychotics, and while there is some suggestive evidence for that from international comparisons, I don't think anyone can assert it confidently. Psychosis can be so disabling and unpleasant that desperate measures providing short-term relief are justified. Some people can tolerate antipsychotic medications for the long term and appear to do well on them, but most people can't. The side effects are just intolerable. The new antipsychotics, which some people can take who can't take the older ones due to psychological and motor side effects, cause extreme weight gain, hyperlipidemia and diabetes, which are life threatening.

So, there is a movement which sees these drugs as little more than a fraudulent conspiracy by their manufacturers. I'll tell you what I think just for the heck of it but I won't spend the time today to justify it. My personal conclusion is that these drugs should be a last resort, not a first resort as they are now.

We certainly should not start feeding people antipychotic drugs at the first signs of psychosis, but rather wait to see if the episode remits. They should never be given to children under any circumstances. People who do start taking them might try cutting down or stopping when they reach their forties, and see what happens.

About antidepressants, I have even stronger opinions. They should be used very seldom. There is no such thing as the disease of "social anxiety disorder," and nobody should ever be given a drug for that diagnosis. If there is a "disease" of depression, it exists in only a minority of people who are diagnosed with it. People who suffer from depressed mood and other symptoms such as overwhelming guilt or apathy should try proven methods such as physical exercise, counseling and cognitive-behavioral therapy. Oftentimes the problem can be overcome by changing one's habits and one's thinking, or taking steps to change one's situation in life. People who turn out to have intractable, disabling depression, might elect to try antidepressant drugs. But they have to understand that their problem is not caused by a "chemical imbalance," nor will the drugs fix one.