What helps, what doesn't

Miscellaneous findings from this week's news:

Eccinacea doesn't help colds.

Circumcision helps reduce spread of AIDS.

Prayer doesn't help heart patients.

Antidepressants may not help most people.

NY's smoking ban helps reduce second-hand smoke symptoms.

Comments

[Karen M]

Elderberry is better than echinacea. Buy Sambucol.

[mudkitty]

Milk Thistle.

[Eli (creepandblink)]

I usually just cure myself by believing really hard that I'm doing something to cure myself.

[janet]

One of the things I wonder about with these antidepressant studies is just what they're looking at -- how they're evaluating the effect of the antidepressant.

For years I was very resistant to the idea of taking antidepressants. I finally decided to give them a try a couple of years ago. Maybe it's a placebo effect that's made the subtle but very real and beneficial change in my mood, but if so it's a placebo effect that's worked continuously for almost 2 years. And this was a period during which I was going through some very difficult experiences. After the first couple of weeks on the drug I had no negative side effects. My emotional range hasn't been blunted at all; I just don't get stuck in depressive loops any more. I also suspect that the drug is what led to an effortless 15-pound weight loss over a period of about a year (not everyone wants to lose weight, but in my case it was a nice side benefit).

I'm not going to name the medication I'm on, because I really don't want this post to be taken as a plug for the product. I was prepared for the medication not to work, and I was also prepared for the possibility that I wouldn't be able to tolerate it even if it did work for me. But in fact the drug has been much more beneficial than I expected, which is another reason that the possibility that it's all a placebo effect seems unlikely.

Anyway, given my experience I really wonder about these studies. Or maybe I've just had unusually good luck with the drug.

[cerebrocrat]

The antidepressant study, fwiw, was a metanalysis and not an empirical study. The statements to the press made by one of the authors certainly suggests an ideological bent on her part. So, I'd want to discuss these authors' methods with a good statistician before getting too excited about it. I don't know whether antidepressant drugs work, but the suggestion that depression isn't a neuropathological state is simply wrong.

[Lindsay Beyerstein]

I don't know whether antidepressant drugs work, but the suggestion that depression isn't a neuropathological state is simply wrong.

Agreed.

FWIW, I believe that most antidepressants have therapeutic value, but needless to say, only for people who are actually in the neuropathological state(s) that these drugs are designed to reverse.

Sadly, like most profitable pharmaceuticals, there's a concerted effort to convince doctors and patients that the underlying pathology is even more widespread that it really is. Which is not to say that depression isn't a common disorder, not by any means.

[rob helpychalk]

Is there a way to define "neuropathological state" that doesn't commit you to naive Cartesian dualism?

I ask because in the popular conception, people with clinical depression have a "chemical imbalance", a physical ailment. They are constrasted with people who "just need to cheer up", who have some sort of moral or character flaw.

So the popular conception of depression depends on physicalizing what can be treated medically, while holding fast to the idea that there is something distinct and mental to moralize about.

So how do you distinguish between the people with a "neuropathological state" that the "drug was designed for" from other states without falling into Cartesian dualism?

[Cookie]

Second on the elderberry. Elderberries make a fab cordial or wine, as well. Much tastier than echinacea.

[Jedmunds]

That anti-depressant article is backward. The more interesting phenomena is why there is such a concerted effort to discredit the very existence of depression? I don't know exactly what's behind it. But it's the same game those Intelligent Design types are playing. And headlining that particular article "Antidepressant efficacy may be overblown-experts" is the same as titling an article "Evolution may be bullshit - experts" and then leading off with something by Michael Behe and then tucked way at the bottom the repudiation by legit evolutionary biologists.

[cerebrocrat]

So how do you distinguish between the people with a "neuropathological state" that the "drug was designed for" from other states without falling into Cartesian dualism?

I see it as a cost/benefit analysis. The relationship between brain and behavior is commonly misapprehended in much the same way that the "nature/nurture" (non)dichotomy is. They're two phases of the same, single system. Clinical depression is a positive feedback loop in which maladaptive behaviors exacerbate a neuropathology which exacerbates maladaptive behaviors, and so on. It's probably true that most depressions could be overcome with only behavioral interventions, and it's also probably true that changing behavior rather than changing biochemistry directly is the best long-term solution to depression. But changing learned behaviors is really, really hard especially in adults. The real value of drug and other interventions (e.g., electroconvulsive therapy) is to bring the biochemical context far enough along toward a nonpathological state so that behavioral inteventions can be applied at a reasonable cost (in patience and expertise as well as money). What separates people who "really need" drugs from those who don't is probably a matter of how far off they are from being able to make necessary behavioral changes. Which is the key to why there is often so much judgement associated with depression.

I'm sure that Lindsey's right that antidepressants are overprescribed, that's just how things are in our system, and I'm also sure that drug + behavioral therapy is always better than drug therapy alone, and that behavioral interventions would often suffice without drugs. But the antidepressant-dissing crowd of the "we should be changing society, not drugging our brains" variety always get on my nerves, because while they're stumping for TV-turnoff week, lots of depressed people are sitting shut-in at home imagining death. Begrudging those folks something that would make them feel better is depraved.

[otherpaul]

Rob and cerebrocrat: is this a job for the hardware/software analogy? If a computer malfunctions, you can fix the problem by making changes to the hardware, but often you can fix the problem, if you know how the program works, by inputting the right commands. The fact that you can 'cure' the thing by talking to it (in its software language) doesn't mean there's a Cartesian soul in the computer or 'ghost in the machine'. It just means that you can attack the problem at more than one level.

Still, I'm a fan of anti-depressants just given my own experience with them. Why bother with all that fancy, behavioural stuff if you can feel better by taking a couple of tablets every morning? Unlike in the computer case, here the hardware intervention is so much easier.

[garnet]

That's a good point about using antidepressants to get a person into a state where behavioral therapies can be effective. The question then is what behavioral therapies. Unfortunately, as far as I know there's no consensus in the medical community about which behavioral therapies are effective; what is recommended is a matter of what your clinician thinks works and what's available in your community. I am personally convinced that the two behavioral therapies that have the most evidence of effectively treating depression are a) regular exercise and b) cognitive behavioral therapy; but the criteria for recommending these over other types of talk therapy, or hypnosis, or acupunture, is even looser and less specific than the process by which a particular person gets matched with a particular medication. And of course it's largely influenced by the local medical culture wherever you happen to be; who the local therapists are; how, when, and where they were trained; etc. It's quite a muddle.

[Herbert Browne]

Re "..While there were no significant changes in upper-respiratory symptoms like coughing.." (from the quoted 2nd-hand smoke study)-
This is probably about the air that's been modified by internal combustion. The rest of the study was quite positive.
Cold coming on? A Chinese OTC herbal called "gan mao ling" has worked better than the old echinacea/ goldenseal/ skullcap remedy did (for about 15 years, now). Also, taking the water-soluble vitamins (esp Bs) before bedtime (instead of pissing them away with the morning beverage flush) is also helpful... ^..^

[otherpaul]

What's the point of behavioural therapy if not to make the person feel better? Assuming we're not non-physical spirits, it should be in principle possible to achieve the same result pharmacologically. In fact, I think some of the current meds can do this. So why bother with some clumsy, time-consuming behavioural therapy?

[Herbert Browne]

"..Assuming we're not non-physical spirits, it should be in principle possible to achieve the same result pharmacologically. In fact, I think some of the current meds can do this. So why bother with some clumsy, time-consuming behavioural therapy?"
(Posted by: otherpaul)
As social beings (whether you can weigh and measure that fact or not) we DO respond to stimulus as evanescent as a conversation. The language we use can also change our internal chemistry (eg laughter as "the best medicine"), as can the weightless images provided by pornographers. All therapies have their limitations... ^..^

[otherpaul]

Herbert,

I agree that language, e.g., can change internal chemistry (just as commands in the right language can change the charges on a computer's board), and that therapies based on that can work. But any given effect can be produced by more than one kind of cause -- e.g., this same internal chemical change can be effected via more than one causal path, and if it's a chemical change, one way of causing it would amount to taking the right pill.

This wouldn't be a good long-term solution if the depression were caused by environmental factors and/or one's behavioural habits. There, one would be better off getting at the route cause of the internal chemical glitch, by addressing these environmental or behavioural factors. But if the depression came first (and led to these behav., etc. problems), and is itself an internal chemical imbalance, then the most efficient way of getting at the route cause would amount to taking the right pill.

[MFB]

Anecdotally, my experience with antidepressants was that they didn't work, whereas psychotherapy did. This doesn't mean "antidepressants don't work, psychotherapy does", of course. I'm me, and you're you.

However, my other experience with antidepressants is that shrinks and quacks prescribe them like Smarties (chocolate sweets in these parts, that is) and don't ask too many questions about how appropriate they are or how well they work. Or exactly why people are getting depressed.

I think it's a good thing to be a trifle sceptical about medication. "Trifle" meaning "keep checking" rather than "throw the damned stuff out".

[mudkitty]

For me, years of psychotherapy didn't work, but 10 months on a low dose of Efexor did, almost right away. That was 5 years ago. Just lucky, I guess.

[Kriston Capps]

Bah. Echinacea leaves may not cure colds, but I have no doubt whatsoever that steamed echinacea extract and lime juice does.

[Alon Levy]

About circumcision, I'd say sewing vaginas reduces the risk of male-to-female transmission of AIDS to almost zero. Hell, cutting off vaginas is even better... I'm not sure whether castration will work, because on the oen hand the penis is still there, but on the other there's no semen.

In related news, cutting off people's arms reduces the chance that they'll commit murder by about 100%.

[Njorl]

One problem is that depression is sometimes just a symptom of something worse. In that case, successfully treating it with either therapy or medication is a disservice to the patient. I was depressed for years. After a particularly bad stretch of life, I was really miserable. I was given a prescription for an SSRI, and it worked on the depression very well. Therapy might have also worked as well, but I don't know.

The problem was that my thyroid gland had nearly ceased functioning. That was the principle cause of the depression. The successful treatment for depression covered this up. It wasn't until the other symptoms of hypothyroid became severe that my doctor would even consider testing me for hypothyroid.

So, because I was not receiving treatment for hypothyroid, I was at a significantly higher risk for stroke and heart attack for about 2 years. Neither happened, I think.

While drug companies and medical insurers might very well be justified in pushing medication in lieu of therapy, there is a danger that they are also squeezing out diagnoses of significant maladies that have depression as a symptom. Therapy proponents pushing back would just exacerbate the problem, as therapy is just as likely to fix the symptom without fixing the problem.

[Njorl]

"Bah. Echinacea leaves may not cure colds, but I have no doubt whatsoever that steamed echinacea extract and lime juice does."

This reminds me of "Granny's Cold Cure" from "The Beverly Hillbillies". She mixed up this vile concoction, and everyone was convinced it cured colds. Yep, just take the potion, drink plenty of liquids, get lots of rest, and in 7-10 days your cold will be gone. Worked every time.

[janet]

My experience was that psychotherapy was worse than useless. Firing my in competent therapist really gave me a lift, though.

[Lindsay Beyerstein]

Rob, I don't think you have to be a dualist to believe that clinical depression is a neuropathological state. I agree with D'Amasio that emotions have functions. They're supposed to provide meaningful feedback to guide rational action.

In clinical depression, emotions become warped. Someone who feels sad all the time because she has a terrible job or a bad marriage isn't necessarily depressed, although chronic emotional turmoil can cause (or coexist with) depression.

I believe that there are major physiological differences between healthy emotional distress and depression. The somatic symptoms of chronic melancholic depression suggests a distinctive neuropathological syndrome. Some patients with clinical depression exhibit symptoms like early morning awakening, dramatically increased REM sleep, appetite changes, psychomotor slowing, and altered patterns of cortisol secretion.

Depression is like chronic pain. Pain is healthy to the extent that it gives us information about physical threats or injuries that we need to protect in order for them to heal. Chronic pain is neuropathological because the phenomenon persists and self-perpetuates, even after the appropriate trigger has subsided.

[rob helpy-chalk]

Lindsay,

I think you are right, but there's more to be figured out. Emotions do have functions, and we can easily identify a class of people whose depressive reaction is malfunctioning. I'm not sure the somatic signs are a good indicator of that, though. Once you start talking function, there is actually a value judgment to be made: is this depression a rational, proportional reaction to the world? The key is not whether one has sleep disruption, but whether what one is going through is worth losing sleep over.

But there's more to it than that. Sure, we can identify a group of people whose depression is the result of a malfunctioning emotional system. (I'm intentionally avoiding saying "emotional module.") But do we want to restrict pharmaceutical therapy to this group? What about people with chronic problems in their lives, the sort of problem that it would be rational to respond to with depression, and whose life problems simply aren't going to get better any time soon. Example: people providing long term care for an elderly dying parent. This sort of situation sucks. It can last up to 20 years. It is reasonable to respond to it depressively. But if drugs can help, why not help. (oh wait, here's a better example: the elderly dying parent themselves.)

ok, I've spent too much time on the interweb today.