The Power of the Placebo
I was visiting my mother in the hospital a few months ago – nothing serious; she reacted badly to anesthesia during an outpatient procedure – and I happened to glance over at the rolling cart they put the meals on that extends over the bed. There was a tube of HeadOn. (For those of you unfamiliar with HeadOn, it’s a tube of wax that you rub directly on your forehead to relieve headache pain.)
“Mom? HeadOn? Really?”
“Don’t start. I know it doesn’t work, but it works for me.”
Well. There you go. The power of the placebo illustrated in one short sentence. It effectively stopped the conversation. Could I argue against it? Should I? I just changed the subject.
A placebo, as defined for clinical trials of medicine, is an inert version of the medicine tested. It allows the experimenter to “blind” the patient, and the person who administers the drugs, from whether the patient is receiving the drug or not. This allows the experimenter to know that any effect would come from the drug itself, not whether the patient knows he’s getting the drug.
Placebo testing can lead to an ethical dilemma, if the medicine is the sole cure for the disease or condition. You can’t use a placebo to set a bone or remove a tumor; that would just be wrong. In the case of pain relief, however, it can be useful – precisely because it has been shown that self-reported results, such as lessening of pain, can largely depend on the patient’s thoughts and feelings.
So in my mother’s case, she knew intellectually that the stick of wax really has no way of deadening pain nerve endings, or attacking the cause of the headache, but she also knew that when she used it, she felt better. It could be, at least in part, the feeling of control she gained over something going on in her body, with the added benefit of no side effects.
I’ve known people who have used magnets in the same way. Their chronic pain was not helped by the medicine their doctors prescribed, and the medicine had very real side effects. The magnets seemed to help, and there were no side effects. So, based on their own experience, they have a very powerful bias against any information that shows that magnets could not help. And in the case of self-reported pain, I always return to that same question: Should we push science-based medicine in this case?
Dr. Harriet Hall, in today’s Science-Based Medicine blog, asks the same question regarding acupuncture and chronic prostatitis/chronic pelvic pain syndrome: “Since medical science has little to offer for CP/CPPS, is recommending acupuncture ethically justified; and if so, should patients be told it is evidence-based?” Should these patients be given placebos and told they are not placebos? Studies have shown that these patients will report less pain, even though there is no real reason for the lessening of the pain.
What’s the answer? I don’t really know, but it’s obvious what my answer has been so far. When faced with the statement “I know it doesn’t work, but it works for me”, my response has been “So…how’s the Christmas shopping going?”
BubbaRich
December 9, 2009 @ 1:51 am
I deal with it in exactly the same way. Well, I’ve started to answer more honestly and directly when well-meaning people try to suggest or give Airborne to me. But I try not to even subtly disagree when people tell me about something they are using that is completely free from any actual effects other than placebo. Wow, I hope nobody gets lost in that sentence…
Anyway, I try to strike a balance. I think it’s definitely morally negative to destroy a cure for somebody, it’s like telling Arthur Dent that he can’t really fly. But if a person is spending substantial money on a placebo, and (wittingly or not) advertising the placebo as an effective cure to cause others to spend substantial money on it…then I might say something. I don’t want people to get rich and make other people poor with this kind of fraud. Well, I don’t actually mind the people getting rich, since that’s morally the same as any sort of entertainment, but I don’t want them to make people poor with it.
Morally, this particular struggle will be eliminated if we can find effective but honest ways to intentionally harness these mechanisms, say some sort of behavioral therapy or meditation. There are already a lot of techniques that work for a lot of people, some of them even honestly. I studied Tai Chi Chuan when we lived in Alabama, and I found huge benefits from it. There were several people in my class who were very intense and evangelical about the woo aspects of it. But I enjoyed my huge benefits without and despite their ideas. Or sometimes I found I could get even more benefits by using their woo as a useful metaphor, say to visualize the qi flowing as I maintained my balance in some poses. But back to the subject, we need to find or create a mechanism or framework of mechanisms to reliably harness the placebo effect for most people for most needs. Sort of a step-by-step way to try different things to find something that works for you, as Tai Chi worked for me, as a real tool getting real, physical mechanisms in my body to work for me.
ReedE
December 9, 2009 @ 11:22 am
Great post Jerry!
I’d thought the value of placebo went away once one saw the wizard behind the curtain, seeing it was just a sugar pill. That the effect persists to some extent is fascinating.
Regarding Bubba’s points, it’s an interesting question where one draws the line.
Where there are no practical alternatives for a patient (such as one who believes that only acupuncture will reduce her pain) we should at least press to make that placebo as safe as possible.
Jerry Jobe
December 9, 2009 @ 1:06 pm
Thanks for the comments, guys.
BubbaRich, I think your experience with Tai Chi is a great analogy – especially since my mom has also taken Tai Chi classes. She never bought into the woo aspects, but they helped her with her balance and strength, which are beginning to deteriorate now that she’s in her seventies. I wasn’t about to talk her out of it, once I realized that she didn’t get into the qi and the meridians, etc.
Of course, I would speak up for two reasons – first, if the placebo took the place of treatment that really works on a treatable condition; and second, if the placebo could actually cause harm, as ReedE points out. I put chiropractic into that second category.
I hope to extend into discussion of that second category in my next post, sometime in the next few weeks.
BubbaRich
December 9, 2009 @ 1:26 pm
Jerry, I meant to add those two reasons as reasons I would challenge these things, but they seem to have stayed out of my post. I used to write notes to remember details like that, maybe I need to go back to that habit at 2am…
Stephen K
December 9, 2009 @ 1:32 pm
Have you seen this yet?
http://www.cosmosmagazine.com/news/3187/mechanism-placebo-effect-discovered
Jerry Jobe
December 9, 2009 @ 1:50 pm
Stephen, I saw that, but I’m going to have to re-read it a few times. The words make sense, but the message has to sink in for a while. Right now it just seems like it’s saying our perceptions can actually affect neurons in our spinal cord. Is this ground-breaking? Widely applicable? I still have a lot of questions that may actually be answered in the article.
BubbaRich
December 9, 2009 @ 5:06 pm
I haven’t read it, yet, but I’m gonna call “BS!” if they’re really saying they have “the” placebo effect…Okay, I’ve skimmed it. This sentence seems to show a confusion about statistics: “When a substantial number of people taking the placebo get better too, it makes it hard for researchers to determine whether a new drug is actually of any benefit.” What it shows is an equivalence between the inactive placebo and the drug being tested (depending on the exact numbers).
This article fails to mention the fairly critical point about whether this study was double-blinded. I’m gonna guess that this study doesn’t in fact show anything new neurologically, just ties it in to placebos, because I can remember asking this exact question in neuroscience class 3.5 years ago and having a detailed discussion with the professor about how pain signals are gated.
I might dig the study up to see whether it was double-blinded, but I’m working through another one right now about nonmaterialist neuroscience (a peer-reviewed paper!), that seems to be fairly bad science.
Stephen K
December 10, 2009 @ 8:19 am
I’m not sure you could actually double blind this study, because there were no active ingredients at all:
“Throughout the study, the researchers applied painful heat to the arms of 15 healthy men, and compared the spinal cord responses when they thought they had been treated with either an anaesthetic cream or a placebo.
Both creams, in fact, were inactive. But the fMRI scans (functional magnetic resonance imaging) showed nerve activity was reduced significantly when subjects believed they were getting the anaesthetic.”
BubbaRich
December 10, 2009 @ 9:05 am
Yep, Stephen, that’s what I was talking about. The people interacting with the subjects should definitely not know that both creams are inactive, for exactly the same reason that they also shouldn’t know which cream is being applied. Although in this case, leaving which cream unblinded would open up something else for an interesting test, and I get the feeling that they might not have blinded that information even from the subjects. It would make sense, anyway.
Stephen K
December 10, 2009 @ 10:12 am
Although, it is possible that the people interacting with the subjects didn’t know, because it does say that the subjects did think some of the creams had active ingredients. I can only get to the abstract, and none of the other articles say anything about the blinding, but it has been published in two different peer reviewed journals, which leads me to assume that it was blinded until I hear otherwise.
BubbaRich
December 10, 2009 @ 1:35 pm
Okay, I’ve downloaded and read the article and supplemental material from _Science_. Here’s what people knew beforehand:
“Neurobiologically, placebo analgesia is in many cases opioid-dependent and relies on frontal cortical areas and their projections to downstream effectors in the brainstem (1, 2). One possible mechanism of placebo analgesia is thus that cortical areas recruit the opioidergic descending pain control system in the brainstem (3), which ultimately inhibits nociceptive processing in the dorsal horn of the spinal cord in a gate-control manner (4). Behavioral data support the idea that placebo analgesia can act at the level of the spinal cord (5), but there is no direct evidence that nociceptive responses in the spinal cord are reduced under placebo analgesia.”
Here’s what they tested:
“We combined high-resolution functional magnetic resonance imaging (fMRI) of the human cervical spinal cord with a robust placebo analgesia paradigm (6) (fig. S1) to test the hypothesis that spinal cord blood oxygen level–dependent (BOLD) responses related to painful heat stimulation are reduced under placebo analgesia.”
They used heat to induce pain sensation in the subject’s arm. They found the strongest spinal fMRI response where they expected it, near the back of the spine between two neck vertebrae. They tricked the subjects into thinking that they had a fake cream and a lidocaine cream, and subjects reported less pain on the “lidocaine” (placebo) side.
Here’s the results:
“Our data provide direct evidence that psychological
factors can influence nociceptive processing
at the earliest stage of the central nervous system,
namely the dorsal horn of the spinal cord. They also
reveal that one mechanism of placebo analgesia is
inhibition of spinal cord nociceptive processing,
possibly mediated by the descending pain control
system (3) in a gate-control manner (4). It is likely
that the decreased BOLD responses we observed
are caused by endogenous opioids because opioid
antagonists block placebo analgesia (1) and because
recent fMRI data from rat spinal cord
showed morphine depression of dorsal horn BOLD
responses (7).”
Here’s their caveat emptor (can you do italics in a response?):
“However, our study cannot reveal the exact mechanism of spinal inhibition [i.e., effects on primary afferents (presynaptic), interneurons, or projection neurons (postsynaptic)] and whether the observed effect is specific for nociception, because we did not measure responses to innocuous stimuli.”
And here is their conclusion:
“Nevertheless, the demonstration that modulatory influences on nociceptive spinal cord activity are measurable by fMRI in humans opens up new avenues for assessing the efficacy and possible site of action of new treatments for various forms of pain, including chronic pain.”
That gives a path to investigate one of the things I was talking about above, a way to reliably access the placebo effect, although only for pain control. The Cosmos writer wants to extend this to “understanding _the_ placebo effect,” but there’s no reason to think that spinal pain pathways are involved in coordinating and facilitating all of the other biological responses that change healing, etc. A friend of mine is researching something that may be related to this, the sympathetic and parasympathetic nervous systems, that also have a spinal portion, and affect heart rate, enzyme production from most internal organs, speed of response, etc. Placebo pain management is important (since we have known for a long time that pain perception is very person- and situation-dependent), but it isn’t “THE placebo effect” that Alissa Jenkins is selling it as.
I can send the article to anybody who is interested, along with the supplementary material. The article was a one-page “Brevia” from Science (Science. 2009 Oct 16;326(5951):404.), and the supplemental material is just 11 (mostly text) pages. The supplement describes the experiment in more detail, and shows that the investigator almost certainly was not blinded, because they talk about the techniques they used to deceive the subjects. Intentional deception changes the need for blinding, and part of what they are measuring is the ability of the investigator to deceive the person into thinking the cream is lidocaine.