Placebos: Threat or Menace?

Feb 22, 2010 16:36

So there's an interesting dilemma been brewing in medicine for some while now, concerning placebos and the use of the placebo effect in clinical practice. A recent study indicates that the strength of the placebo effect can be augmented by "a supportive psychosocial context". Now, the study is for patients with irritable bowel syndrome, a ( Read more... )

Leave a comment

Comments 36

meihua February 22 2010, 21:38:23 UTC
Wasn't there some research which indicated that placebos (and nocebos) still have some (though less) effect when the person taking them knows they contain no active ingredient?

Reply

randysmith February 22 2010, 21:46:03 UTC
This.

In more detail, for me "not lying" trumps "better effect" except in extremis, but I believe that some of the placebo effect can be gotten without lying, if you put some thought into it. Heck, even if it requires belief on the part of the patient, there's some room for working here--get in some "faith healing" of the type the patient believes in to augment whatever the main medical treatment is. For me it'd be Reiki or energy work, for a Christian laying on of hands by a priest/minister, etc.

Another thought (that I'd be curious as to your response to): A while ago, a friend of mine commented to me "Given how big an effect placebos have, why isn't a hypnotist part of the standard battery of specialists I have attending on me (on par with anesthesiologist, surgeon, etc.) when I go into surgery?"

Reply

meihua February 22 2010, 21:49:29 UTC
This, back at you! :)

The medical profession and its regulators should be looking seriously at some of the reasons why patients are going to "natural health" practictioners or what have you.

Here's the first reason - when they're getting reiki, they're not a patient, they're a person, working with a professional and a person, together on their problem. And that person has more than seven minutes to speak with them. And the environment is quiet, and peaceful, and perhaps there's music on. And during the treatment, there's listening, and a real sense of connection.

Of course, at the same time, the reiki practictioner is waving their hands around and doing pretend magic. But what if a real doctor, doing real doctory things, also did all of the above, save the magic?

Reply

docorion February 22 2010, 22:04:33 UTC
I object to the first reason entirely. I treat people, also. A reiki practitioner may not refer to someone as a patient, but a patient they are nevertheless, and every alternative health practitioner I've ever been to (and I've been to many, although few reiki practitioners) has approached the problem as the person with the greater knowledge *of their specific discipline*. If you are seeing a doctor who does not "work together with you on your problem", fire them. Simple, straightforward. (Also, if you have ever been to an emergency department which is quiet and has music, tell me *now*; I will leave here to work there **so fast ( ... )

Reply


volta February 22 2010, 21:48:26 UTC
I am not entirely convinced it is lying to the patient to give them a pill and say "this should help you", even if there are no active drugs in the pill. You would not be issuing the placebo if you did not think it had a chance of helping them, right?

Reply

docorion February 22 2010, 22:10:39 UTC
Depends on the setting and the medical problem. And also-most people are less comfortable nowadays taking pills for which the only explanation is "This should help you". The concept of "antibiotics" and ":narcotics" and the idea that meds have principles of action is quite widespread, even in the scientifically challenged. So most folks want at least a brief treatise on *why* the drug works, and won't take it if *they* don't believe it will work.

I do this by talking about the way the drugs work in the most basic terms I can find. It mostly works for me; the reason most of my patients cite for not taking the meds I prescribe is cost. (The second most common reason is that it wasn't the Percocet they wanted, that Vicodin shit never works, doc). But if I give a placebo, I need to make something up. Not sure I'm OK with that.

Reply

volta February 22 2010, 23:52:59 UTC
Ah. I am way off-curve here, I guess. I do not think I have ever had a doctor explain what they were prescribing beyond "this should take care of problem x". My general distrust leads me to do my own research rather than asking, if I do not already know what the drug does.

I agree, having to make something up to get them to take it does introduce an ethical question I was not seeing earlier.

Reply


anaisdjuna February 22 2010, 21:48:40 UTC

If they gave something that was a mild painkiller, but wasn't constipatey which can't be good in this sitch... and/or something that was a good supplement for bowel health then they wouldn't be lying... and could still cash in on giving the patient some relief and leveraging the functional augmentation that seems to be a part of it all.

Reply

docorion February 22 2010, 22:15:22 UTC
Yeah, but the widespread use of placebos is a slippery slope, and my colleagues already do a poor job of communicating. See below for more thoughts on this.

Reply

anaisdjuna February 22 2010, 22:35:38 UTC

Yeah. It's a slippery slope alright. Especially in more hardcore instances where people missing out on real treatments. Like the House episode where 13 thinks she's getting better in a drug trial but then Foreman finds out she's on the placebo and switches her.

Everything is a slippery slope it seems....

Reply


koshmom February 22 2010, 22:07:43 UTC
Some patients, when they visit a doctor, expect that doctor to Give Them Something for their illness. If the patient doesn't get any pills, they think the doctor either doesn't know what they have, or is a quack for not curing them with a medicine. Thus, a placebo is a good choice, because the patient gets what they perceive they need, yet the doctor doesn't give them medicine they don't need (and may get addicted to).

Reply

docorion February 22 2010, 22:14:30 UTC
Yeah, I've heard that. I treat that problem with information, which takes longer, but ablates the problem, in time. The reason more docs don't do that is the perception that it takes more time. I don't find it does, but I also talk fast and have canned explanations of why antibiotics/benzodiazepines/narcotics are a Bad Idea in this situation. I am happy to share those with the doctors who are unwilling to make up their own. But in part (IMNSHO), one of the doctor's jobs is to *explain* the disease process. If you give a pill and say "Next patient, please!", you've failed in that duty. That is one of my objections to placebos-too many of my colleagues would use them as substitutes for communication.

Reply

onemintjulep February 23 2010, 14:53:17 UTC
Press Ganey. Metrics. RVUs. Door to Doctor. Ad Nauseum.

Reply


beowabbit February 22 2010, 22:44:01 UTC
I checked all your ticky-boxes, because I’m torn too. I think “don’t lie to your patient, even by omission” trumps almost everything else. On the other hand, if a doctor told me “I know of no pharmacological reason why this thing should affect your disease, and as far as I can tell it’s just a placebo, but in a study eight percent of patients showed quantifiable improvement on it and it’s cheap and it doesn’t interfere with the rest of your treatment,” I’d probably take it.

I also suspect that there are things that we think are placebos that actually have some physiological mechanism behind them, and things that we think are “drugs” whose actual benefit is mostly placebo effect. (In theory, testing for efficacy ought to decrease that, but if the drug gives you a mild tingly sensation and that’s all it does, it might be a more effective placebo than the other placebo.)

Reply


Leave a comment

Up