SSRI antidepressants no better than placebo for most; the myth of mental illness?

A recently published meta-analysis of studies of SSRI antidepressants - including unpublished trials - shows that they are no better than placebos for all but the most severely depressed people; and furthermore, that severely depressed people exhibited a decreased placebo response rather than a increased responsiveness to the drugs. (SSRIs, “selective serotonin reuptake inhibitors” are the class of drugs that include Prozac (fluoxetine),
Effexor (venlafaxine), Serzone (nefazodone), and Seroxat (paroxetine).)

This prompted a discussion over on Slashdot, where I posted as follows:

Depression is not illusionary, it's a real disease.

Depression is certainly not illusionary, and is certainly a real problem.

Whether it is helpful to call depression, and other "problems in living" that are not directly diagnosed as neurological lesions or malfunctions, "diseases" or "illnesses", is questionable. I suggest reading Thomas Szasz's The Myth of Mental Illness (available also in an expanded book form, but the original paper gives the gist of it):

...In actual contemporary social usage, the finding of a mental illness is made by establishing a deviance in behavior from certain psychosocial, ethical, or legal norms. The judgment may be made, as in medicine, by the patient, the physician (psychiatrist), or others. Remedial action, finally, tends to be sought in a therapeutic -- or covertly medical -- framework, thus creating a situation in which psychosocial, ethical, and/or legal deviations are claimed to be correctible by (so-called) medical action. Since medical action is designed to correct only medical deviations, it seems logically absurd to expect that it will help solve problems whose very existence had been defined and established on nonmedical grounds. I think that these considerations may be fruitfully applied to the present use of tranquilizers and, more generally, to what might be expected of drugs of whatever type in regard to the amelioration or solution of problems in human living.

I'm not sure whether I agree with him entirely or not. I do think that "mental illness" is at least in part a social construct - but the same is true of a lot of physical illness.

Telling someone who is suffering from mental illness that the pain they feel is not their fault, but a disease which can be cured, can oftentimes be comforting.

The problem is there are two distinct messages: "it's a disease" and "it's not your fault". The one has nothing to do with the other. For example, if despite knowing full well the risks and having resources to help quit, you smoke three packs a day and get lung cancer, it is a disease and it's your fault. And of course, having a disease doesn't imply that it can be cured.

Also implicit in the disease model is that the treatment comes from outside: take this pill or have this surgery to get fixed. But if these conditions are regarded as "problems of living", this introduces the possibility of the sufferer being the primary problem-solver, of taking positive action (with therapist as guide and assistant) rather than being a passive patient.

Is it a "disease" if someone lacks normal and usual cognitive and behavioral skills to deal with, say, anger? I don't think it's useful to categorize such a lack alongside clear biological dysfunctions like the flu and coronary artery disease. Is it their fault if they lack these skills? No, not unless they've been given full opportunity to acquire them and have chosen not to.

"It's not your fault, you haven't done anything wrong" and "there are solutions, and you can get help to find them" are good messages. Very good messages indeed. But "your condition is a disease" doesn't really encompass either of those.

Dude, just tell me that someone who is clinically depressed or with some other severe mental disorder doesn't have a disease to my face, ok?;)

Well, if it makes you happy, I'll tell you to your face that they may not have a disease, depending on how we define it. I'll tell you that clinical depression is a very different sort of entity than physical ailments, since it is diagnosed based on judgments of behavior rather than physical evaluation.

Depression (the disease under discussion) can be very real.

I thought we'd broadened the topic to general "mental illness" by now. And if you don't think anger and depression are linked, ask around. But look, the reality of depression is not in dispute here; what is being questioned is its categorization.

A broken leg is a real problem but is not a disease. (Indeed, I think that the concept of "mental injury" might serve as decent analogy in many cases, such as PTSD.) Various sorts of ignorance are real problems, can even be fatal (e.g., "only gays and drug users get AIDS, so I'm safe!"), but ignorance is not a disease. Depression is a real problem; many people think that categorizing it as a disease is inaccurate and/or not helpful.

One big reason is the sort of interventions it suggests. Diseases don't get better when you learn new skills; but cognitive-behavioral therapies are quite effective at helping people with depression. Labeling depression a disease automatically puts the emphasis on drug therapies (and, gods forbid, on psychosurgery or shock treatment).

Ever seen a person who used to function fine slowly (chemically) imbalance to where they can't get out of bed?

The "chemical imbalance" hypothesis is one of the big questions here. It's pushed by drug companies that make SSRIs, but there's a lack of evidence for it. (Which is not to say that mental activity doesn't have neurochemical correlates, just that they're not as simple as "low serotonin == depression".)

But yes, I've seen people - friends, family members, a housemate of mind - go into serious states of depression where they became unable to function. I fail to see how calling them "sick" would have helped.

Ever felt it yourself?

Self-diagnosis is a minefield. But I will say that years ago, my doctor dropped several gentle hints that I should consider SSRIs. I'm glad I didn't; instead I've made positive changes in my life, and I'm feeling much better now. Probably the most important involved becoming more aware of my body, learning to break the feedback loops of muscular tension. (Which is not to say that some drugs don't help some people.)

(oh, and you can see brain chemistry changes in autopsies http://www.channel4.com/science/microsites/S/science/body/depression.html, just like hardening arteries, so, yeah, it's a physical disease)

Autopsies of people who committed violent suicide, which should hardly be taken as representative; and your linked article notes that "post mortem analyses are complicated by factors other than depression that may change brain chemistry. The mode of death, previous drug history, current therapy, and time between death and autopsy can all affect the results."

It may well be the case that some people who are diagnosed with "clinical depression" have a neurological disorder. But what is being diagnosed by the term "clinical depression" is not that neurological disorder, but a behavioral pattern.

You may get various diseases as a result of those activities, but a there is no "Alcoholism" disease which has caused you to unwillingly drink yourself into a stupor every night.

Actually, alcohol is one of the few genuine addictions, where continued use causes changes to the nervous system such that the drug must be present for normal functioning. Alcohol withdrawal can kill, so I'd say that's evidence of a genuine disease.

Now, does having a disease mean that your drinking is "unwilling"? That's a whole different question. Certainly it provides a strong motivation - "if I don't drink, I feel very ill, and indeed might die. Bottoms up!"

But alcoholism is quite likely overdiagnosed, a confusion of alcohol abuse with alcohol addiction.

Placebos / drugs work because they're empowering - I'm doing something, I'm taking something, things will get better.

A friend of mine recently remarked that she feels better just having the prescription for some of these drugs. She never even fills them. It's the ultimate in homeopathy!

I heard a tremendously interesting episode of Radiolab a few days ago dealing with the placebo effect, worth checking out.