Jan 27 2013
Is acupuncture as effective as antidepressants? Part 2. Blinding readers who try to get an answer
You are currently browsing comments. If you would like to return to the full story, you can read the full entry here: “Is acupuncture as effective as antidepressants? Part 2. Blinding readers who try to get an answer”.
129 Responses to “Is acupuncture as effective as antidepressants? Part 2. Blinding readers who try to get an answer”
Not that long ago, when open access publishing was just starting off, I looked into publishing in PLoS One due to the great work that had been done (and still is being done) with PLoS Medicine and PLoS Biology. Thankfully something stopped me – and it was not the publication fee… It was the fact that PLoS One explicitly told everyone that the manuscripts would not be reviewed for quality, only methodology and whether the methodology agreed with discussions and conclusions. PLoS One would publish any investigation that is methodologically sound, with the effort to act as a co-financer of the more prestigious journals of PLoS (at that time Medicine and Biology)!
Over the years this has become evident with an increasing number of articles without real merit in PLoS One, especially in the fields of medical interventions. I think we still only see the beginning of “worthless peer-reviewed publications” and the development of a multitiered journal hierarchy. That something can be found in the “scientific literature” has historically been a problem for “grey sources” (unpublished theses, obscure journals, small languages etc), soon it will be a problem of “(semi)predatory publishers”… PLoS needs to rethink their publishing model soon, or else they will be seen as nothing more than a scientific vanity press.
# Exilapotekare, I think you misconstrue my criticism of PLOSOne. I applaud its willingness to publish all methodologically sound articles because this is an important corrective to for profit journals publishing only positive results and leaving on published the results of negative trials. As Ben Goldacre has spelled out in Bad Pharma selective publication in the for profit journals leads to dangerous and ineffective drugs being promoted with citations of publications in high impact journals they convey a distorted view of how they work.
We need to have all results of all trials available in print, but we need some assurance that these results are presented accurately and transparently so that readers can form their own opinions. Such a process will sometimes let some bad science and bad medicine into print, and that is why I am calling for citable post publication peer review.
I have severe treatment-resistant MDD. If it weren’t for this blog and excellent posts like this one, I could easily fall prey to alternative medicine — hell, I’ll try anything at this point — especially since Khan et al conclude, “These data suggest that type of treatment offered is less important than getting depressed patients involved in an active therapeutic program.”
So thank you very much.
P.S. potential COI: I work in evil Big Pharma; also, one med (bupropion) did wonders for me until it stopped working — so I take issue with those who claim that antidepressants are no better than placebo.
I presume the reference is this. It specifies neither the manufacturer nor the wavelength used. This fellow appears to recommend the near-IR, which would be invisible.
(Also: Seriously, PLoS is including MS Word files as supplementary material?)
Been there. Gotta love it when the term of art is in fact “poop-out.”
If I am reading all this correctly, the overriding problem is that you cannot compare these “effect sizes”, contrived as they are out of before-and-after comparisons within single arms of studies of different treatment methods, unless you knew that every detail of the study populations and their management was the same — which you can’t, even when pooled, this being the very reason for the R (randomization) and all the blinding in the usual double-blinded RCT. Without such an insecure footing we cannot say with confidence what this data means..
OTOH, if the effects of antidepressants are, as posited, in substantial part due to placebo and nonspecific influences then it is not clear how better studies could settle the question of relative effectiveness of some of these treatment programs, so as to support sound informed consent.
For example, it would be unethical now to perform a study comparing acupuncture and antidepressants in patients with severe depression. One suicide would be too many. Yet with mild depression the results could probably be easily manipulated either way. It is almost inevitable that at least some studies would favor acupuncture in cost/risk/effectiveness terms especially once drug side effects were taken into account.
How would we react to that? On our answer hangs the justification for our determination to oppose acupuncture under all circumstances.
There remain good reasons for the mainstream choosing not to endorse acupuncture, but they are clearly not based upon the application of a “gold standard” of evidence, once you start to consider “acupuncture” as the sum of a variety of nonspecific (but therapeutic) influences, as many indeed do nowadays.
This statement from the paper thus probably has a substantial grain of truth at least wiht mild cases “These data suggest that type of treatment offered is less important than getting depressed patients involved in an active therapeutic program“
#Pmoran This is an excellent summary and reaction to my blog post and gave me lots of food for thought. You hit the nail on the head in articulating why these authors’ calculation of effect sizes was an accurate and misleading.
A previous paper by one of the authors of the acupuncture/antidepressant paper, Irving Kirsch, is also an author of what is among the most highly cited PLOS Medicine paper so far
http://www.plosmedicine.org/article/info:doi/10.1371/journal.pmed.0050045
and it thoroughly confuses the issue of whether the difference between an active condition and a placebo control group in some way identifies the size of the mechanism of the active treatment. Harriet Hall has blogged about this particular paper and effect sizes, maybe more than once
I think that it is useful to compare an intervention and a pill placebo condition, under appropriate circumstances, to determine the relative efficacy of an intervention. However, that efficacy of an intervention could be due to mechanisms very different than what enables a pill placebo to have an effect. So, unless we have testable ideas about mechanism, we can’t assume that an intervention incorporates the same mechanism as a pill placebo in an additive way. Efficacy of an antidepressant is not necessarily equal to some biologic effect plus the effect observed with pill placebo in any simple additive way.
I’ve had to think a lot about acupuncture as treatment contact with for depression. I think we can agree that there is only a nonsensical claim of any plausible biological mechanism and at least some semblance of a claim of testable mechanisms for psychotherapies. When we get to the clinical level of administering a treatment for depression to an actual patient, we often need to think about adapting the treatment to that patient, particularly when the initial treatment is not having satisfactory effects. Conventional psychotherapies allow a lot of leeway for therapists’ problem-solving. For instance cognitive behavior therapy can be adjusted if a depressed persons actually getting more ruminative when there’s so much focus on their thinking. The focus can sift to behavior activation or collaborative empiricism that gets the depressed person off their butt and out testing their ideas in the real world. That can be accommodated within the theory of mechanism. I don’t know what you would do if a patient is it responding to acupuncture. If you really believe in the mechanism, maybe you could move the needles or poke the person more often, but that gives limited options. What you think about that?
[...] Right Way to Praise Your Kids – webmd.comIs acupuncture as effective as antidepressants? Part 2. Blinding readers who try to get an answer – sciencebasedmedicine.orgDo penalties for smokers and the obese make sense? – [...]
Not really. They are definitely based largely on application of the “gold standard” of evidence, because, as we’ve pointed out time and time again on this blog, the better the quality RCT, the more indistinguishable acupuncture effects are from placebo interventions. It doesn’t matter where you stick the needles, nor does it even matter if you stick the needles in. Add to that the complete lack of anything even resembling a plausible physiological rationale for why acupuncture might “work” for the various conditions for which it is touted, and there is plenty of reason based on a combination of “gold standard” evidence and lack of biological plausibility to dismiss acupuncture as nonsense.
David, We are agreement on acupuncture. But, however ineptly, I was trying to point to the dilemma of someone who advocated it has a treatment, justifying it because all treatments are supposedly equal in depending heavily on a ritual and a rationale. As Khan/Kirsch argue in the PLOS article, even sham acupuncture is suitable.
I don’t buy that, but suppose for the sake of argument, we assume that someone is going to acupuncture for depression and getting treatment, whether sham or otherwise. What Irving Kirsch and a lot of people don’t acknowledge is that many persons have to get some adjustment in what ever they are receiving for depression before they get a benefit. So, treatment can’t be a matter of a rigid routine, even with the rationale. My point is what would an acupuncturist offer under the circumstances? I think they would be stuck moving needles around. Psychotherapists, in contrast, would be guided by their sense of mechanism to try something different than they had been trying.
James Coyne: “Psychotherapists, in contrast, would be guided by their sense of mechanism to try something different than they had been trying.”
And in some cases psychotherapists will admit their limitations and suggest that the patient should go to another therapist working within different framework. I’ve seen it happen from time to time. Or suggest going to a psychiatrist to augument psychotherapy with antidepressants.
Totally agree, #Alia, at least I would hope that they would. Quite difficult to predict ahead of time what evidence-based treatment will work for the individual depressed patient. And therapy + meds on average does better than either alone, particularly with more depressed patients. But you raise an interesting question: if acupuncture is obviously not producing improvement in a patient, where would the practitioner make a referral? Somehow I doubt it would be to a conventional mental health professional. Rather, to some a practitioner of other woo.
Dr. Coyne,
Thank you for the post. It is from posts like these that I learn the most. I am still trying to make sure I fully understand the nuances since I have not had the time to really read it in detail.
The difficulty I see with the application to acupuncture, as pmoran posits, is beyond the fact (as Dr. Gorski has pointed out) that “gold standard” evidence actually does show it to be nothing beyond placebo. Because, as I have argued here before, when it comes to psycotherapy many things can reasonably be considered an “active” intervention. Literally everything about simply being in a study, interacting with all the people involved, etc are actively affecting the pathological organ – the brain. The same cannot be said for actual disease of other organs; acupuncture will not affect the beta cells of your pancreas and fix your diabetes.
The difficulty comes from scratching one micrometer past the veneer of the “therapeutic ritual” that is acupuncture.
A patient comes to me as a physician, and is depressed. I offer:
…pharmacotherapy. “How is that going to help me?” “Well, there is data… neurotransmitter theory of depression… corrects imbalance… has been shown to be reasonably effective” “How does it do that?” “By inhibiting the reuptake of serotonin… etc”
…talk therapy. “How is that going to help me?” “Well, there is data… neural plasticity… working out problems verbally and addressing them can actually help restructure brain pathways… has been shown to be reasonably effective.” “How does it do that?” “Your brain can actually change connections and restructure pathways if you give it the chance… talking through your problems can actually help literally change your brain directly, which is reflected by your slowly changing thoughts and attitudes…”
…acupuncture. “How is that going to help me.” “Well, there is data… which shows that other stuff that works might not be better than acupuncture… and some that shows it can maybe help people who have failed other therapy…” “How does it do that?” “Well, the needles don’t actually do anything… you just have to believe that they will… because really all that is going on is non-specific interactions between you and some nice old Asian man… and that can restructure your neural pathways…”
“…so why are you telling me to get needles stuck in me again?”
Of course, your point about adjusting therapy is just as good, if not better, a reason. But really, you just have to either believe the needles do something yourself or hope that your patient doesn’t ask the simple question of how it works. So you are either incompetent or unethical. Because the moment you admit it has nothing to do with needles, you are stuck trying to justify how that is a better option than a day at the spa, a concert, or actual psychotherapy. Or you have to tell your patient that it will realign his meridians to allow the chi to flow properly and adjust his attitude. Which means it doesn’t matter if it is an acupuncturist who will refer you to a naturopath or a physician doing the acupuncture. It’s just a non-starter.
Tomorrow, when I have some more time and a fresh head, I want to re-read and delve more deeply into your critiques as a learning exercise. But I felt it incumbent to point out that if the rationale for acupuncture in depression is the ancillary patient-practitioner interaction, then psychotherapy is a much better and more ethical option, despite the fact that I would agree that the non-specific aspects of acupuncture in this case actually are active interventions directly treating the disease process (though in a roundabout way via lying – by ommission or otherwise – about the nature of acupuncture).
James Coyne: What Irving Kirsch and a lot of people don’t acknowledge is that many persons have to get some adjustment in what ever they are receiving for depression before they get a benefit.
I am not sure how you know that. Are there studies? It may merely mean that “tincture of time” or unknown incidental happenings in the patient’s life have at last worked some magic.
# nybgrus I’m not sure I could’ve reticulated your points better than you did and I certainly agree with them.
Having almost obsessively read this PLOS one article in order to provide the detailed critique I have across these two blog posts, I’m convinced that these authors had no particular confidence in acupuncture as a treatment for depression. Rather, they were trying to stay on message in making the point that one of them, Irving Kirsch, has made so often before, that all that we do for depression is simply a matter of placebo. Unfortunately in the case of this PLOS one article, they had to really torture the data that really didn’t support this point.
#Pmoran evidence-based psychotherapy for depression is highly manuallzed and a therapist who is not achieving the intended result can simply follow the manual and shift gears.
Practice guidelines for antidepressant treatment of depression recognize the patient should be showing improvement at five weeks and if there is no improvement, some action should be taken. Actions might involve encouraging adherence with patient education, adjusting dosages, or shifting medication. Unfortunately, in primary care treatment of depression, where most treatment occurs, there often isn’t an assessment at five weeks and so overall outcomes are quite poor and substantially worse than what is achieved in clinical trials.
This is a great deconstruction of an article, and a great investigation into what really helps for depression.
I want to add two aspects that others are addressing, but not in the way I as a psychologist see them.
I have been trained in a very mainstream, normal way, including with an eye toward evidence. We psychologists (some of us – there are a lot of off-the-map psychologists out there) recognize that one factor that works fairly reliably for depression is “behavioral activation.” Doing something rather than nothing.
With a behavioral approach, specific behaviors would be determined to counter specific aspects of the depressive experience, but overall a certain level of activity would be encouraged, and also “pleasurable experiences” would be encouraged, since often a depressed person has a notable imbalance in daily life of “unpleasurable” to “pleasurable” experiences. It is nice to prescribe someone to enjoy an ice cream per day, or a relaxing bath per day.
“Behavioral activation” is the concept in my mind when I see “exercise,” weight-lifting,” and even “bright light therapy.” Doing something is better than nothing, being proactive may boost self-efficacy, and other various modest effects may be involved, such as social involvement. The effect may be a “fake it til you make it” effect, with self-efficacy and increased decent social interaction experienced as a side effect of increased activity.
So, hopefully this gives others a concept for seeing the plausibility of a varied range of interventions that otherwise seem pretty random, and seem like they are interventions somehow parallel to therapy or medications. I just see them as “behavioral activation.”
But I also mention this to destroy the idea that these are “alternative therapies.” They are components of a decent cognitive-behavioral intervention. While exercise is not itself a “psychotherapy,” your counselor should be assessing your daily activities and interactions, and figuring out where you might be lacking, and making it a priority to convince you to get active, to encourage/push/motivate, and to problem-solve your failure to get this going. That is part of decent psychotherapy for depression: the therapist should be helping you fake like you are in the swing of things, until you do get the normal drive or desire back to do this. It is a real boot-strapping thing.
Also, my second comment is that one man’s placebo effect is another man’s “nonspecific therapeutic factors.”
Irwin Yalom’s book on group psychotherapy is the classic source to grasp the influence of “nonspecific therapeutic factors” as active ingredients in therapeutic change. Yalom was considering group psychotherapy, so he does not discuss “depression” so directly as he discusses progress and change generally.
You can probably google “nonspecific therapeutic factors” to find the list, but they include things such as “instillation of hope” and “universality.” In clients who begin to improve, these factors are endorsed as being perceived by the client. Those not improving don’t perceive these phenomena.
Along with Yalom’s nonspecific therapeutic factors, Carl Rogers came to the opinion that “unconditional positive regard” is a necessary condition for improvement. Again, he did not focus so closely on the medical model concpet of “depression,” but addressed a range of mental health difficulties. Unconditional positive regard means that the therapist, no matter what, has a favorable opinion of the client, and conveys this simply by a decent, respectful interactional style. This includes our listening skills: nodding and saying “um-hummm” to show you are following a discussion, parroting, reflection, summarizing, posture, etc., are the techniques or ingredients that sum up to convey the attitude of “unconditional positive regard.” you can google that, also.
For us psychotherapists, these “nonspecific factors” are bread and butter therapeutic phenomena.
Outside of psychotherapy, others don’t quite conceptualize these in this way, and they are comprehended in a vague way, and labelled “placebo effects.”
By theory and evidence, change happens by the practice of specific (identifying and challenging depressogenic thought patterns, behavioral activation, relaxation, autogenic training, etc.) and nonspecific factors (mentioned just above.
These ingredients will often work even if not delivered in a psychotherapeutic context. Why would they not? If they work, they work. Hence, a support group or good friend can provide a certain dose of these.
When a client goes to an acupuncture session, or goes through the detection-and-intervention process for antidepressant medication from GP or psychiatrist, plenty of these nonspecific factors are involved.
So, I encourage people to grasp these concepts and learn about them. “Behavioral activation” and “nonspecific therapeutic factors.” I believe it will help you see what is happening when a person shows a good response to acupuncture or placebo drug, or “autogenic training.”
As an example, this helps me understand how acupressure can yield the same degree of effect as acupuncture, and ends up not being such a good “control” condition for an acupuncture trial.
JC, thanks for all the work on this topic.
“Blinding” readers is funny: the quadruple blind – pt, provider, assessor, and study reader.
MedvsTherapy – If I had the option, I would “like” your comment.
Me too, and yours too, # MedsVsTherapy.An update on the effects of this pair of blog posts.
It has stimulated a lot of exchange within the huge editorial board of PLOS One, with all of the responses highly positive and supportive. It is generally acknowledged that my critique exposed an article that had escaped adequate peer review. But the dilemma is that detailed commentaries such as I provided in my blog posts do not yet have a citable, indexable in ISI web of science, outlet in PLOS one. Everyone seems to be admitting that that’s a problem, but has someone said, changing that policy of a huge Journal with tens of thousands of reviewers and over 15,000 papers published last year, is like trying to change the course of the Titanic.
I’ve been invited to do a third posting on this article at the PLOS blog Mind the Brain that specifically takes the perspective of a bewildered depressed person trying to choose between psychotherapy and acupuncture on the basis of what they read in the scientific literature. Stay tuned…
MedVstherapy, As an example, this helps me understand how acupressure can yield the same degree of effect as acupuncture, and ends up not being such a good “control” condition for an acupuncture trial.
Yes! Those studies are not truly testing efficacy, and certainly not overall effectiveness within practical medicine. They are testing out aspects of Chinese medical theory and practice that we would normally not bother considering, regarding them as having no important, or unique, physiological significance or plausibility.
Nevertheless, in SBM-speak the failure of minimal variants to obviously outperform each other somehow morphs into a bald “‘it’ doesn’t work!”. What is it that isn’t working, when we are also obtaining those unusual effect sizes from such variants?
Nybgrus: The difficulty I see with the application to acupuncture, as pmoran posits, is beyond the fact (as Dr. Gorski has pointed out) that “gold standard” evidence actually does show it to be nothing beyond placebo.
You may not have noticed that Dr Gorski’s response was not to the point. The critical context-setting ending to this sentence of mine (highlighted below) was cut out out.
There remain good reasons for the mainstream choosing not to endorse acupuncture, but they are clearly not based upon the application of a “gold standard” of evidence, once you start to consider “acupuncture” as the sum of a variety of nonspecific (but therapeutic) influences, as many indeed do nowadays.
For one thing, as is now made clear, there is no body of quality head-to-head comparisons to go on. It is not even quite clear what “quality evidence” would consist of in such a study, siince either side could easily boost reported outcomes by maximizing various non-specific influences and their accompanying element of reporting bias.
The answer probably is that there is no answer at present, and there probably won’t be until or unless we can somehow focus upon more objectively measurable outcomes, or develop objective surrogates for true treatment efficacy.
I don’t think I get this. The publication strategy is already haphazard. The addition of citable rebuttals only represents another layer. Does Thomson Reuters even care?
I’m probably going to regret getting into this again so I will do me best to be brief.
You realize, pmoran, that you are now in essence arguing that there is no data that we can actually parse to answer the question of acupuncture and the “non-specific therapeutic ritual” of it when we just had it out about how you claim there are piles of data to support these “surprisingly large effect sizes” from acupuncture.
So which is it? Do we have data demonstrating these effect sizes or do we not?
You then go on to say that what we have is essentially a giant hodge podge of data from studies that aren’t designed to give us the info we are seeking on the topic, so we can’t really draw our “skeptically dogmatic” conclusions… yet you can somehow use that to infer utility of acupuncture?
And you completely gloss over the fact that the vast majority of research on acupuncture is done by people who actually are looking at the purported putative mechanisms of it… not “as the sum of a variety of nonspecific (but therapeutic) influences, as many indeed do nowadays.” It is we skeptics who are the ones seeing it as the sum of a variety of nonspecific influences. Not the people actually studying it!
#Narad. With respect to original research, the publication strategy of PLOS One is coherent and represents a strong corrective to confirmatory bias and selective availability of positive studies in conventional print journals. Studies have to methodologically adequate and transparent about their limitations. This cannot occur perfectly and thus the necessity of citable, indexable corrective commentaries.
http://www.plosone.org/static/reviewerGuidelines#criteria
Criteria for Publication
To be accepted for publication in PLOS ONE, research articles must satisfy the following criteria:
The study presents the results of primary scientific research.
Results reported have not been published elsewhere.
Experiments, statistics, and other analyses are performed to a high technical standard and are described in sufficient detail.
Conclusions are presented in an appropriate fashion and are supported by the data.
The article is presented in an intelligible fashion and is written in standard English.
The research meets all applicable standards for the ethics of experimentation and research integrity.
The article adheres to appropriate reporting guidelines and community standards for data availability.
Yes, but I said haphazard, not incoherent; the intended target was breadth, which goes to the size of the “editorial board.”
Again, I get this. What I’m not getting is the actual obstacle to implementing the goal. I take it that such commentaries, if indexed as part of the journal, should be interpretable as being up to the standards embodied by the journal. Is this the intersection, in that the model doesn’t have a method to apply an imprimatur to a response that the input gantlet doesn’t expect to see?
Nybgrus: I’m probably going to regret getting into this again —
No, you’ll be fine, eventually. It’s simply a very difficult notion to us to entertain with our conventional medical background and its (largely understandable) biases, that, to make the point quite clear, a strong collection of placebo and other non-specific influences could match weakly active pharmaceuticals when caring for patients with certain illnesses.
But it could be true. There already studies suggesting this possibility (including the present rather flawed one), but for the reasons that I have mentioned it is difficult to interpret them. We do need to be able to respond to them in a more sensible way than referring to those rather irrelevant differences between nominally “real” and “sham” acupuncture.
This is uncharrted territory for all of us.
Most of the recent posts on this forum seem to miss the fact that most of the “placebo” response in antidepressant trials does not come from a real placebo effect, but from spontaneous remission. Depression is by its very nature a cyclical disorder. Most people will eventually remit spontaneously even without treatment, and the goal of antidepressants is to speed remission and to prevent relapse later on. There have been trials where they compare people in the “placebo” arm of a trial to people who are on a waiting list to be enrolled in a trial, and they’ve seen indistinguishable remission rates between the two groups. This isn’t something that’s been studied well enough for us to know how much of the effect in the placebo arm is due to nonspecific therapeutic effects vs. spontaneous remission, but the data we do have suggests that spontaneous remission is the vast majority of it. There’s an underlying narrative that depression is particularly susceptible to the placebo effect because it’s “all in your head” or somehow not a real disease, but it’s more likely due to the fact that it’s simply cyclical.
The second thing is that antidepressant trials are extremely suspect because it’s become nearly impossible to recruit a suitable patient population. It used to be that there were lots of depressed people who were untreated, but now the vast majority of them are treated by their primary care provider, so whoever you have left is treatment-refractory, or more likely, an unemployed person who will say anything or feign any symptoms to get a check. A lot of these people are enrolled in 5-6 clinical trials simultaneously. It’s hard not to be too cynical, but the studies we have are generally of such low quality (or studying things in such an unrealistic way) that it’s hard to make treatment guidelines based off of them.
p.s. the most compelling evidence of antidepressant efficacy is not in treatment of acute depression, but in prevention of relapse. Those studies show that people who stay on antidepressants vs people who are tapered off to placebo have something like three-fold lower relapse rates. The evidence of SSRIs in anxiety disorders is also more compelling. So Kirsch and other antidepressant denialists are playing a game similar to creationists who keep talking about some weak spot in the fossil record while ignoring all of the more compelling genomic data.
evilrobotoxxx:
Thanks for that. I think I knew that somewhere in my head, but it wasn’t something at the forefront of my conscious thoughts on the matter. I think it is quite helpful to augment my thoughts on this topic.
pmoran:
I don’t know what you are on about. I’ve even said it again in this very thread – I have absolutely no reservations about so-called placebo having a direct, tangible, and measurable benefit in depression (and other psychiatric disorders to varying degrees). I also explained why I think it reasonable to consider them active treatments and they are mislabled as placebo effect, except insofar as that we lump everything that is not the dependent and independent variables into that category.
And I’ve also said it innumerably – if I have a bias it should be in the favor of CAM and placebo. I was educated to believe they were both valid constructs and powerful. It was after doing ever more reading in depth that I was forced to modify my outlook to be in line with the evidence.
Now the only leg you may have to stand on regarding this particular topic is if the nonspecific effects of acupuncture/pressure/magic joo joo are actually more consistently beneficial and/or with a much larger effect size than psychotherapy. However, there is absolutely no a priori rationale for this to be the case and zero evidence to support the notion. So it is very unscientific to insist strict agnosticism towards the topic – we can at best say we are agnostic but it is unlikely. And even if it were true, then it would be our goal to tease out why and remove that from the BS that is acupuncture itself. Because as I said in my first comment here, at some point in some way you must be lying to your patient in order to treat their depression with acupuncture. So no matter the effect size or consistency demonstrated at some future time, our hands are bound.
@evilrobotxoxo – good comment too! I’m curious, would you say then, based on the other information we have – wait list compared to placebo – that the benefits found from accupunture are more likely due to cyclic remission than nonspecific therapeutic influences?
@MTR: yes, I’m saying that the “placebo” arm should be called the nonspecific effect arm. Nonspecific effects include spontaneous remission (usually referred to as “regression to the mean” for conditions that aren’t necessarily cyclical), nonspecific therapeutic influences (i.e. the “placebo effect”), dishonesty/unconscious bias (patients and researchers artificially inflating a patient’s symptoms to get them enrolled in the trial, then artificially deflating them out of a desire to see people get better), etc. The problem with the concept of the “placebo arm” of a trial is that it implies that the apparent treatment responses in untreated people are due to the “placebo effect,” when in reality they’re due to a bunch of different factors.
This is Arif Khan, MD, the primary author of this manuscript. I am happy to note that this manuscript has generated discussion which as you all know is the point of scientific publication. First and foremost, my intention (in part representing all of the authors) was to present depression research data, a journey analogous to the one taken by Alice in her wonderland. There are a lot of advocates and believers, but relatively few critical thinkers.
Having said I would like to make two further points. Specifically, conducting a meta-analytical synthesis is not easy or the data adequate to come to firm conclusions. However, these data represent a common sense and clinical problem that physicians and allied health professionals face everyday. Here is my version that I couldn’t put in a manuscript, but can address in a setting like this.
Of every hundred human beings who develop the depressive syndromes, about 10 will ‘improve’ with time alone, the so-called waiting list, spontaneous remission etc being the descriptors.
Of the 90 patients left, about 20 to 30 will get better whether they see an African shaman, Indian Sadhu or get fake accupuncture. There are several descriptors that people have used, with of course placebo being the most misused and abused word. As many of you on this website have stated it is the non-specific therapeutic effect.
Besides these so-called controls, it is likely that another 5 to 10 will ‘improve’ if they see somebody who has more of an expertise with mood disorders. Specifically, forms of psychotherapy and antidepressants. For both of these, you have to pay the piper. Psychotherapy can go wrong and is expensive and time consuming. Antidepressants are simple, foster magical thinking, are easy to take, although you are left with the side effect burden, which luckily in my own life time has changed for the better.
Further intensity and intrusion such as combined medication and psychotherapy, medication combinations, light therapy, ECT, TMS all will all enhance the proportion of those doing better by another ten or so.
So, here is the crux of the problem. If you have added right, we still have more than 40 of the original group of depressives not ‘improving’.
The take home message from me as the author and clinician is that our understanding and treatment of depression is not as good as it should be. At best, the emperor is partly clothed, not naked as some would like to point out. But, hopefully, we can get him clothed and in better shape, but this is only possible if we appreciate what is missing. Arif Khan, MD
Has anyone ever considered here, or has any scientific research been presented to show that possibly a change in a depressed persons individual circumstances may be most effective. You know… having a more fulfilling job, more money to pay the bills, a better relationship or even a nicer society to live in. Has any research on these issues been done? Or are we stuck with defining this as a medical condition to be ‘cured’ by pills, therapy, acupuncture or placebo?
The problem is, it’s not always the case of individual circumstances. Take my sister – she had a fulfilling, well-paid job, interesting hobbies, some close and not-so-close friends with whom she could talk and have fun, support in her family (that’s me). And depression. She went through group therapy, which helped her a lot and after 10 years I hope we can safely assume that her episode of depression is over.
Or take a friend of mine – she’s been at least mildly depressed throughout her life, always sad, never felt true joy. All that was until she went to a psychiatrist who found a right antidepressant for her. A few months ago she said something like “Now I finally know what it’s like to feel happy”.
So I guess we can sum it up “It’s complicated”. Yes, in some cases a change in individual circumstances will help. In some it won’t.
Alia says ‘Take my sister – she had a fulfilling, well-paid job, interesting hobbies, some close and not-so-close friends with whom she could talk and have fun, support in her family (that’s me).’
That is your view. What was hers?
Hers was unhappy love afair. And OK, maybe that’s individual circumstance – but you must admit that it’s sometimes hard to find someone to happily fall in love, just like that. And barring that, group therapy was the second best thing for her.
Research such as somehow “prescribing” these items?
@Dirk Stool
As a trained (but not practicing) therapist,, I can say there actually has been at least some research dedicated to this topic. The ones that spring to mind are in regards to income….while there’s evidence that incomes over certain amounts do not do much to increase happiness ($45k/year for one person is the number if I remember correctly), amounts below that most certainly increase depression risk (can’t find the study now, I’ll have to look for it later).
As for a nicer society:
http://www.ncbi.nlm.nih.gov/pubmed/23349294
I think some of the problems with treatment come in when either 1. there’s no contributing problem the patient can point to or 2. the contributing problem is unsolvable (or very difficult) or 3. an underlying mental illness makes a lousy life circumstance harder to deal with. I dealt a lot with #2 when I ran grief groups through a hospice. The loss of a loved one was not fixable, but we did work hard to help people find appropriate resources to help them deal with the practical matters around their loss before referring them to a doctor to talk about medication.
It must be noted though that (at least in my state) a master’s level therapist who does not inform a patient that medication is available and could help can be found negligent and liable for damages if the patient later goes on medication and finds it works. I think this duty to inform can sometimes be construed by people as “pushing” meds…..and of course there are some providers on all levels who do actually push meds more than I think they should.
I’m sure I can dig up more studies on the topic if you’re interested.
I remember the first time I sought help from a therapist/psychiatrist who prescribed the Paxil I was taking and CBT. I talked to my brother afterwards. He said “How can you be depressed. You have a good job you love, live in a great town. It sounds like you have great friends…what could be so wrong?” I said, “Yeah, that’s kinda the point. I’ve got all that, but I still feel bad.”
I’m sure life circumstances can send people into depression, In fact I know it’s happened to me. But sometimes things just change, hormones-something in your brain- whatever and you feel very different about basically the same circumstances. Kinda hard to explain to someone who hasn’t experienced it.
The thing is, assuming that your feelings are always dependent upon circumstances can be very problematic, because you keep trying to fix something that’s basically not broken to make yourself feel better. Sometime, the reality is, your life is fine, your brain (hormones, whatever) need work.
evilrobotxoxo:Most of the recent posts on this forum seem to miss the fact that most of the “placebo” response in antidepressant trials does not come from a real placebo effect, but from spontaneous remission
That ‘s true. Spontaneous remissions will indeed tend to dampen any differences found when treatments are compared and it is one obvious factor in why so many different treatments seem to have about the same order of “effect” upon depression. I suspect that the SR rate is considerably more than Arif’s 10% in the population that the usual family doctor will be asked to treat.
However, as you go on to allow, there should be similar rates of spontaneous remission in a waiting list or “usual care” group. In most studies where comparisons are made to those, complex, patient-involving and ongoing interventions like acpuncture outperform those groups by a substantial margin (0.5 effect sizes) — and with many conditions, not only with depression.
Yet –
There have been trials where they compare people in the “placebo” arm of a trial to people who are on a waiting list to be enrolled in a trial, and they’ve seen indistinguishable remission rates between the two groups
Presumably those studies used a pill-placebo, which may merely have performed poorly under the conditions of those particular studies. Wouldn’t most anti-depressant researchers do their best to reduce placebo responses?
The present question concerns more active, involving and complex interventions in an almost certainly more promising therapeutic environment.
I state again that have no particular brief for acupuncture, and I have nothing against antidepressants (– well, some tiny nagging suspicions, perhaps — but who wouldn’t have those?). I simply feel we should be being more careful in what we are saying on certain matters. As Arif implies, we cannot yet be quite sure what our own evidence means.
So I see no grounds for changing usual medical practices, but grounds for less hostility to other claims, for now, when treatments are safe.
Alia
‘Hers was unhappy love afair. And OK, maybe that’s individual circumstance – but you must admit that it’s sometimes hard to find someone to happily fall in love, just like that. And barring that, group therapy was the second best thing for her.’
Richard P Bentall has continually pointed out that therapy – any therapy where the ‘patient’ believes someone is actually caring will cause an improvement in depression. This is why acupunture may work. This is why antidepressants may work. But I do not believe it. Improvement in personal circumstances will work and will result in less relapse. How this condition (depression) has become a ‘medical’ problem and not a socialogical problem is beyond belief to me. Maybe you can convince me otherwise.
# mousethatroared
‘I’m sure life circumstances can send people into depression, In fact I know it’s happened to me. But sometimes things just change, hormones-something in your brain- whatever and you feel very different about basically the same circumstances.’
Maybe. I am only asking to see the science behind this. Where is it?
# bs king
“As a trained (but not practicing) therapist,, I can say there actually has been at least some research dedicated to this topic. The ones that spring to mind are in regards to income….while there’s evidence that incomes over certain amounts do not do much to increase happiness ($45k/year for one person is the number if I remember correctly), amounts below that most certainly increase depression risk (can’t find the study now, I’ll have to look for it later).2
Yes I am aware of this study. But not sure I understand your point here. So everyone who does not earn $45k per year is suffering from a medical condition called depression to be cured by pills? Do you think this?
Dirk Stool – Oh well, I haven’t yet seen your evidence that depression is caused by life circumstances, either. Why is that your default?
Are you assuming that ALL mental disorders are caused by life circumstances or do you think that depression is distinctly different from anxiety disorders, schizophrenia, bipolar, etc?
How do you propose implementing this? Take two bootstraps and call me in the morning?
A second response
Dirk Stool “Maybe. I am only asking to see the science behind this. Where is it?”
I think it’s published in books and journals and stuff…maybe you should start in the library.
I would just like to point out that the ‘placebo effect’ only reports on the subjective experience of the patient. So for depression the HAM-D checklist is used. There may be many reasons (Milgram Experiment etc) why people may present false information to a doctor. There is also a report comparing patients reports to the efficacy of ECT to the doctor and another person which shows major discrepancies. I actually do not believe in the placebo effect.. mind over matter?.. based on such research as this http://europepmc.org/abstract/MED/8942043/reload=0;jsessionid=hgCStEFaCEitO6FEBwe9.2. I could cite other research that reaches similar conclusions. I think a further blog is required here rather than just random comments…
# mousethatroared
‘Dirk Stool – Oh well, I haven’t yet seen your evidence that depression is caused by life circumstances, either.’
OK. So you think depression is not caused by life’s circumstances but by something else. I am not sure I agree with you here. What do you think the ‘else’ is? Evidence would be welcome. Of course I know that depression can be caused by well known medical conditions. Thyroidism for example. But a ‘mental disorder’ I am not so certain about.
“Are you assuming that ALL mental disorders are caused by life circumstances or do you think that depression is distinctly different from anxiety disorders, schizophrenia, bipolar, etc?”
There are almost 400 mental disorders (DSM5). Do you think all of these are NOT caused by life’s circumstances? Where is your evidence. Can you cite some scientific research please.
# Narad says
‘Improvement in personal circumstances will work and will result in less relapse.
How do you propose implementing this? Take two bootstraps and call me in the morning?’
I am only suggesting that the best way to effect a cure is first and foremost an understanding of the cause. Maybe our only ‘cure’ is to tranquillise. Do you support this method?
Anyway I do not have answers. I just want to get the questions right first.
(Sigh)
Answer the original question.
Wow, I went down to the coffee shop and this thread really lengthened. Not to slight some of the comments that have been made, but I would like to return to Dr. Khan who so graciously joined the discussion.
#arkhan
Thank you Dr. Khan for offering your opinions about African shaman, Indian Sadhu, fake acupuncture, and antidepressants. However, in two blog posts and the accompanying threads, we have been discussing your systematic review that appeared in PLOS One. Because it is labeled a systematic review, we would expect that a systematic methodology been applied to arrive at some evidence-based conclusions. However, once I looked at the review more closely, I see that it concluded with a set of opinions more akin to what you are expressing here then to what the data supported.
I confess to not being able to follow the journey through the literature with your systematic review as a guide, because it’s lack of transparency as to just what was done and the links between what was done and your conclusions was, to borrow an apt phrase from you, “a journey analogous to the one taken by Alice in her wonderland.”
Let’s start with the acupuncture literature. When I went to systematic reviews available in the literature, I came to a very different appraisal than the one offered in your review. The impression I got was that the acupuncture literature was largely untrustworthy and that with reference to depression, the low quality studies did not afford any evaluation of its efficacy. One narrative review even raise doubts as to whether acupuncture trials described as randomized really were randomized, in the conventional Western sense.
It was very difficult to see how you selected five of the many studies of acupuncture for depression available, but three were from a single American group and two were from China. All three of the papers expressed reservations about being accepted as anything but preliminary or exploratory. None of them provided any head-to-head comparisons between acupuncture and antidepressants or acupuncture and psychotherapy. I don’t see where you felt confident in making conclusions about these issues in the absence of such comparisons.
When it came to extracting, summarizing and integrating data, you applied an approach that I’ve never seen outside of an article co-authored by Irving Kirsch, who happened to be a co-author in this instance. Your approach is novel and I would appreciate you showing me where anyone else has used it besides Kirsch and co-authors. Having done meta-analyses before, you are undoubtedly familiar with the pitfalls and problems in integrating data from diverse studies. There are tools for determining whether the studies are too heterogeneous in designs and results to attempt an integration, but of course those tools are quite fallible. However I see no evidence that you attempted such a check on what you did. As I pointed out in my blog post, and please correct me if I am wrong, the approach that you took destroyed all benefit of the data coming from randomized trials. To get to a more concrete level, what basis do you have for confidence that your comparison of a sham acupuncture group with antidepressants can legitimately be made? Can we even conclude that similar patients are being studied in acupuncture studies, psychotherapy studies, and evaluations of antidepressants?
I could go on and on about the oddities of your systematic review and its departure from established standards and its conduct and its reporting. But let’s start with these questions.
Again thank you very much for joining the dialogue.
Narad
” I am only suggesting that the best way to effect a cure is first and foremost an understanding of the cause. Maybe our only ‘cure’ is to tranquillise. Do you support this method?
Answer the original question.”
Sorry. Which was? I did say ”Anyway I do not have answers. I just want to get the questions right first.”
@mousethatroared
Why do you sigh?
Dirk Stool – I’m glad that you don’t think you have the answers. Neither do I. But if you think that anti-depressants and Cognitive Behavior Therapy (The two most evidence based treatments for depression, I believe) are akin the tranquilizing, then it seems that you have some incorrect preconceived notions.
Personally, as a patient, I feel somewhat at a loss. Do you expect someone to deliver you neuropsychology 101 on a blog? I’m afraid I don’t understand where you are coming from.
There probably is some sort of general blog on the science behind psychiatric disorders…somewhere. If you are genuinely interested, not just interested in arguing an agenda, maybe someone will be kind enough to post a good link. I’m not sure that I know one off the top of my head.
Um, the one in what you were replying to? Allow me to jog your memory. You asserted the following: “Improvement in personal circumstances will work and will result in less relapse.” I asked you how you thought this would be implemented.
No, you directly advanced an “answer.”
@mousethatroared
Dirk Stool “Maybe. I am only asking to see the science behind this. Where is it?”
I think it’s published in books and journals and stuff…maybe you should start in the library.
Yeh but yeh but yeh.. I quite like this guy and what he says..
http://www.youtube.com/watch?v=JKRIoY9aXK8
Maybe you can suggest the library I need to read. Can you link? Thanks.
And by the way, I never said that depression isn’t caused by life circumstances. Please review my comment.
@Narad
The question?
“Improvement in personal circumstances will work and will result in less relapse.” I asked you how you thought this would be implemented.
I did say ”Anyway I do not have answers. I just want to get the questions right first.”
No, you directly advanced an “answer.”
It is my personal belief that improvements in social, political, and personal circumstances can cure the ‘disease’ of depression. I do not know for sure . What is your evidence against this view?
I suppose my only question is ‘is depression a medical disease caused by a biochemical/genetic/physiological abnormality. Or is it caused by adverse human conditions. The answer determines the treatment and the cure. I donot know. Do you?
@ mousethatroared says
“And by the way, I never said that depression isn’t caused by life circumstances. Please review my comment.”
I never said you did.
Look. I have been banned by Steven Novella from his blog http://theness.com/neurologicablog/. I expect to be banned from here. Just getting my questions in first. Thanks.
Dirk Stool “# Dirk Stoolon 30 Jan 2013 at 6:26 pm
“I suppose my only question is ‘is depression a medical disease caused by a biochemical/genetic/physiological abnormality. Or is it caused by adverse human conditions.”
Why do you think it’s either/or? Does the answer depend upon the patient or is all depression exactly the same?
http://theness.com/neurologicablog/index.php/responding-to-a-szaszian/
Bye bye.
@mousethatroared
Dirk Stool “# Dirk Stool
“I suppose my only question is ‘is depression a medical disease caused by a biochemical/genetic/physiological abnormality. Or is it caused by adverse human conditions.”
Why do you think it’s either/or? Does the answer depend upon the patient or is all depression exactly the same?
How do you tell what is what ? Scienctific evidence? Please cite your sources. Thanks.
Please someone cite the scientific paper that proves that depression is not caused by personal circumstances. That is all I ask. Thanks.
Okay then. Thanks Dirk
Please someone cite the scientific paper that proves that depression is not caused by personal circumstances. That is all I ask. Thanks.
well?
http://www.madinamerica.com/
http://theness.com/neurologicablog/index.php/responding-to-a-szaszian/
Burning books is a easy answer. What is the question?
I am sorry to have ever questioned the diagnosis.
Color me unsurprised.
I thought 42 was the answer.
Same. It was not immediately obvious the trollery in Dirk but nearly immediately. I’m happy to continue to ignore. Especially considering that Dr. Khan has joined the discussion and that is vastly more interesting and educational.
Well, you could have clued me in… but luckily Dirk was kind enough to do so himself.
pmoran said,
“a strong collection of placebo and other non-specific influences could match weakly active pharmaceuticals when caring for patients with certain illnesses.”
Dr. Khan said,
“Of the 90 patients left, about 20 to 30 will get better whether they see an African shaman, Indian Sadhu or get fake accupuncture. There are several descriptors that people have used, with of course placebo be
ing the most misused and abused word. As many of you on this website have stated it is the non-specific therapeutic effect.”
–And I will revisit my point, and add a point that I guess was not totally obvious:
There can be two categories of influence working upon depressed enrollees in a depression trial: one is genuine, recognized placebo factors, and the other is nonspecific therapeutic factors.
I entered a relatively LONG post to illustrate what nonspecific therapeutic factors are. They are actual psychotherapy curative mechanisms. I try to “activate” or “use” these when I am counseling.
Before I greet you in the waiting room and ask you to come back to the counseling room, I take a moment and put my life’s concerns and worries to the side, and I mentally dedicate my next hor of life to you. I think about how God loves you unconditionally, and that I am instructed to love my enemy since my enemy is actually simply just another of God’s children, and I reflect on the fact that emotions, thoughts, and behaviors all seem reasonable once you can see the world through the client’s point of view.
I then remind myself of the many people that have gotten better from therapy, and have effusively provided me thanks – then I believe things can get better, and I am in a mental state to impart hope – not pollyanna but realistic, seen-it-happen hope.
I then take a deep breath and go greet you warmly in the waiting room. I am prepped to maximize the nonspecific therapeutic factors.
That is my ritual.
I give you no ritual other than general pro forma. I don’t use lingo, and I challenge you to avoid the lingo that others may have thrown on you. for example, I always claim that I do not know what “bipolar” means, what “narrative” means, and what “hyperactive” or “ADD” means.
I make you break it down into actual descriptions.
I never promise or swear something will work. I know that most of my interventions will work about 2/3 of the time, and fail the rest. I am honest. When you hear this, you figure out that I am honest and real. You are getting unconditional positive regard.
When I express uncertainty, but still seem to be optimistic, you feel hope.
These are active ingredients of psychotherapy. The unspecific therapeutic factors.
When I identifiy maladaptive thought patterns and help you develop more realistic ones, that is a specific therapeutic factor.
When I teach you to practice deep breathing and pratice thinking positive thoughts when in cetain difficult situations, those are specific therapeutic factors.
Placebo effects are totally different from nonspecific therapeutic factors.
In placebo-prone situations, you usually have a great set-up for both nonspecific therapeutic factors and placebo factors.
I believe both happen. I believe both are separate phenomena.
A placebo effect can clearly be suspected when a dubious person takes a biologically inert intervention and gets relief – it happens. There is something about our brain and physiology that makes this belief-based effect happen. But that is not a “non-specific therapeutic factor.”
Placebos can do amazing things well beyond psychotherapy. Drop blood pressure, etc. Don’t confuse the two influences.
A therapist seeking to boost response by a placebo effect is doshonest. A therapist seeking to maximize non-specific therapeutic factors is just an educated and strategic practitioner.
On a different note: Feyerebend and others are big on noting how science cannot exist outside of culture / cannot be conducted outside of culture.
Scientific standards grew and evolve through culture. That is what we are doing here.
Part of the prevailing culture is that an article is published, and it stands, nearly always unchanged, with any corrigendum much less disseminated than the original paper.
There is nothing in science that declares that results must appear in a peer-reviewed journal.
This is a social convention. A decent but not optimal strategy for sharing findings, which is a tenet of science.
Coyne brings up a great limit in our prevailing culture of science: how do we provide alternative points of views, and rebuttals, in a platform coensurate with that of the original publication?
A paper is published, and has some great weight over decent criticisms. That ship cannot turn quickly.
i have learned so much at SBM and at Bad Medicine (same web-identity). It is sad that this type of learning has not been well-recognized and harnessed by the prevailing science culture. Also, NYT commenters can be amazingly insightful. It is like these college classes we took where the prof wanted to throw a topic out there, and get intellectual work out of a class discussion. Usually these are lame. But on SBM, or Bad Med, or some NYT articles, you get tons of insightful, relevant discussion.
I have many articles saved and put aside in folders b/c some anonymous fake-name blog commenter posted their two cents.
The best term for this is crowd-sourcing. You throw an issue or matter out ot a crowd, and let them address it.
This criticism and comment is an essential component of science. But it happens to an almost negligible degree via the “peer-reviewed journal” mechanism.
I agree with JC (being a Christian, how could I not disagree with a JC) that a journal article should be both a finished product and a conversation starter.
I believe that this could be in the future. Once all journals go electric, the ability to post a comment or rebuttal will be there. Comments may need to be in more than one category. Simple voting up-or-down by posting campaigns cannot swamp genuine discourse. Would moderation be OK? sure, except for when the moderator silences relevant points. So, maybe more than one type of response posting would be needed.
SBM could delete all of the alt-medicine posters, who are usually shown to be plain ol’ cult members. However, if those comments were routinely policed and deleted, a fair bunch of mine would have been deleted, and if wrong I at least can post references to support the reasonableness of my points.
The alt-med true believers posting here are, in the long run, a strength. They could over-run SBM, however, and a bunch of fools could over-run an online peer-reviewed journal that had a “comments” section for every manuscript posted.
So, I don’t know how things could be done better, but I agree with JC that the prevailing model of joournal-based peer-reviewed article falls short when measured by scientific ideals of opennness, scrutiny, and debate.
“I suppose my only question is ‘is depression a medical disease caused by a biochemical/genetic/physiological abnormality. Or is it caused by adverse human conditions.”
Sorry to jump in so late. I was taught that if winning the lotto cured your depression, then you weren’t really depressed.
This observation, of course, is utterly devoid of semantic value.
“I suppose my only question is ‘is depression a medical disease caused by a biochemical/genetic/physiological abnormality. Or is it caused by adverse human conditions. ”
Where is the scientific research that backs up the claim that depression is caused by a chemical abnormality and not by circumstance?
Of course even by asking this question I am accused of a being a troll. It seems to me that in the US Bad Pharma adverts have conditioned thinking to such an extent that my thoughts have become heretical. And of course no-one in the USA understands irony. Is there a pill for this also?
So my proposal for advancing science is; take a bunch of depressed people. Give a third of them $50,000 each, give a third prozac, and give a third acupuncture/placebo. After 4 weeks measure their depression using the Ham-D scale. Publish the results.
What would you expect to find?
As the Buddha once asked ‘How do we cure the suffering of the human condition?’ His answer was ‘right thoughts’ Your answer here is ‘right chemicals’ . You lot are following a religious belief and not a scientific quest.
When I suffered from MDD I went to my doctor and he prescribed me 4 drams of whisky a day. It really helped and I swear by it! I have already put myself forward for the trials by David Nutt in the UK to see if MDMA or Ketamine will assist me also. I can’t wait….
Oh! Did I forget to mention the medical marijuana? It is a buzz man! I immediately forget my depression caused by the fact that I cannot pay the mortgage and that my ex wife is a bitch milking me for all she can get.
Did I discuss the fact that my psychiatrist has suggested that I have an overactive amygdala. He suggests ECT as a cure – http://theness.com/neurologicablog/index.php/how-electroconvulsive-therapy-works/. If that does not work we are going to investigate surgery. I am looking forward to it.
I am called a troll just because I advance my opinions. You lot make me feel depressed.
What pill do you recommend?
and acupuncture and prozac has not helped not one little bit. Nor did the placebo.
When I was in the mental hospital I met a woman being treated for severe depression. She lived in a two bedroom house with her new partner and her three sons. She told me that her psychiatrist had recommended ECT to cure her woes. I laughed. My psychiatrist wrote that this reaction was the result of my having a chemical imbalance that would be cured by neuroleptic drugs. So I took them and they were not bad at all! Apart from I now have this incurable twitch affecting my mouth – tardive dyskinesia – i think he said. . But my psychiatrist has told me that they have a drug that can cure this also! Thank god I have insurance and thus reassurance. That put my mind at ease.
One has got to laugh yeh? Only my psychiatrist doesn’t like it when I do. So I have to suppress it. So now I am diagnosed with MDD.
I know we take banning very seriously here, but Dirk has already mentioned being banned at NeuroLogica and linked the post where Dr. Novella discussed him and he is obviously just spamming the comments at this point. I’d much, much, much, rather see a discussion betwwen Dr. Coyne and Dr. Khan if that is even a possibility at the moment.
I am trying to make a serious point also. Although my use of humour and irony is ignored. Let the debate begin. I will disappear now.
I second that motion.
Spam is only good in a Monty Python skit.
Dirk is not banned—yet. He’ll get one last chance before that happens. However, because of his flooding this comment thread with drivel, thus annoying regulars and drowning out substantive discussion of Dr. Coyne’s post, Dirk’s comments will from this point on all be moderated (until I decide otherwise), with all the attendant delays in appearing. Any attempt at sock puppetry on his part will result in immediate banning with extreme prejudice.
Perhaps you may address the issues that I raise? Before you ban me.
Of course I know that my views are heretical. And will result in a ban. I have studied myth and reality. I know we can dismiss the old myths as being pathetic. I try to disclose the myths that we currently live by. A taboo subject. Ban me. I am used to it – by your skeptical religion.
I am a supporter of the views of Karl Popper. I know that these views are not in fashion now. But I will try to promote them still. Ban me. This action says more about you then it does about me.
Oh David Gorski! You recently provided a link to mental illness denial on rationalwiki… http://rationalwiki.org/wiki/Mental_illness_denial
are you really trying to dismiss the anti-psychiatry movement with this link. I urge everyone to read it and decide. Thanks.
Hey David Gorski. Just ban me. Like Steven Novella did. I ask awkward questions which you fail to answer. So ban me.
Do the bAD Pharma contribute financially to this site? How much are you willing to sacrifice your ideals?
of course I know how much money Steven Novella and David Gorski receive from Pharma to promote their views. A simple google search is enough.
Oh dear. My views must be moderate else I am subject to ‘ awaiting moderation’.
Thank you very much David, this my first exposure to the sort of thing. On my other blogs I just get an occasional Scientology associated spamming.
In an effort to get things back on track, I’ll repost a comment, directed to Dr. Khan author of this article we have been discussing, and hopefully get his response.
#arkhan
Thank you Dr. Khan for offering your opinions about African shaman, Indian Sadhu, fake acupuncture, and antidepressants. However, in two blog posts and the accompanying threads, we have been discussing your systematic review that appeared in PLOS One. Because it is labeled a systematic review, we would expect that a systematic methodology been applied to arrive at some evidence-based conclusions. However, once I looked at the review more closely, I see that it concluded with a set of opinions more akin to what you are expressing here then to what the data supported.
I confess to not being able to follow the journey through the literature with your systematic review as a guide, because it’s lack of transparency as to just what was done and the links between what was done and your conclusions was, to borrow an apt phrase from you, “a journey analogous to the one taken by Alice in her wonderland.”
Let’s start with the acupuncture literature. When I went to systematic reviews available in the literature, I came to a very different appraisal than the one offered in your review. The impression I got was that the acupuncture literature was largely untrustworthy and that with reference to depression, the low quality studies did not afford any evaluation of its efficacy. One narrative review even raise doubts as to whether acupuncture trials described as randomized really were randomized, in the conventional Western sense.
It was very difficult to see how you selected five of the many studies of acupuncture for depression available, but three were from a single American group and two were from China. All three of the papers expressed reservations about being accepted as anything but preliminary or exploratory. None of them provided any head-to-head comparisons between acupuncture and antidepressants or acupuncture and psychotherapy. I don’t see where you felt confident in making conclusions about these issues in the absence of such comparisons.
When it came to extracting, summarizing and integrating data, you applied an approach that I’ve never seen outside of an article co-authored by Irving Kirsch, who happened to be a co-author in this instance. Your approach is novel and I would appreciate you showing me where anyone else has used it besides Kirsch and co-authors. Having done meta-analyses before, you are undoubtedly familiar with the pitfalls and problems in integrating data from diverse studies. There are tools for determining whether the studies are too heterogeneous in designs and results to attempt an integration, but of course those tools are quite fallible. However I see no evidence that you attempted such a check on what you did. As I pointed out in my blog post, and please correct me if I am wrong, the approach that you took destroyed all benefit of the data coming from randomized trials. To get to a more concrete level, what basis do you have for confidence that your comparison of a sham acupuncture group with antidepressants can legitimately be made? Can we even conclude that similar patients are being studied in acupuncture studies, psychotherapy studies, and evaluations of antidepressants?
I could go on and on about the oddities of your systematic review and its departure from established standards and its conduct and its reporting. But let’s start with these questions.
Again thank you very much for joining the dialogue.
Hazards of blogging. Dirk is nowhere near the worst. Given his perseveration on this issue, one wonders if he is somehow associated with Scientology. However, sadly, it’s not just Scientologists who promote the sorts of views he does. He’s also confirming the correctness of my decision as we speak; there are several comments by Dirk in moderation that are, well, just like the comments that got him put into automatic moderation.
I used to have a very libertarian “free speech above all else” attitude towards commenting, to the point where I’ve sometimes gotten into arguments about it, but one thing I can’t tolerate is threadjacking and comment flooding, the latter of which is a transparent attempt to drown out discussion through sheer quantity.
No David. I have nothing to do with scientology. Straw man fail.
so you can divert the qustion I ask. What is the scientific evidence to state that depression is the result of disease and not personal circumstance. Please cite the evidence and I will be convinced.
I am quite happy to discuss this issue with Arif Khan, MD,
of course my first question is… how have you defined depression. How have you recruited your subjects?
I think is making a point questioning the effectiveness and safety of a lot of treatments that are considered empirically supported (even in this very discussion it has been noted that much data about such treatments go unpublished), as well as the problems with our psychiatric diagnostic system, on which effectiveness of treatment applications are based. Given the discussion, I think it is a valid perspective to include.
I am also trying to wrap my mind around what seems to me to be an arbitrary distinction between evidence supported psychotherapies for depression and acupuncture for depression in regards to proposed mechanisms of treatment effect and rigidity to treatment manual. Whether or not Kirsch/Khan believe in a mechanism for depression relief for acupuncture, perhaps many acupuncturists do. To my understanding, acupuncturists believe that a lot of distress/illness is related to blockages of flow of body energies, and managing body energy through particular acupuncture based on the patient’s symptoms/experiences may help. Whether or not this is a mechanism or not, many evidence supported psychotherapies don’t have well tested mechanisms for their treatment approach, regardless of the effectiveness of the approach. So I don’t know why the mechanism of treatment effect that I proposed is any less of a “proposed” mechanism than other treatments that only have “proposed” mechanisms of change. Additionally, given such a proposed mechanism, I don’t why it is insisted that in psychotherapies, therapists can make (seemingly radical) deviations from treatment protocols as more information about a patient’s response to treatment is known, but disbelief that this could happen in acupuncture. If you believe you are trying to manage body energy and flow, than if a patient does not respond well or responds differently than expected, than a different (but not inconsistent with proposed mechanism or theory) acupuncture approach would be warranted (focuses on different parts of the body or blockages, increasing/decreasing energy elsewhere, seeking more information from patient about other problem areas in their life, etc.). So an accupuncturist, just a like a therapist, can make response-based changes to approach informed by treatment mechanism, evidence, and theory to hopefully increase the benefit patients experience from treatment
I am also a little confused about to what extent treatment can deviate from protocols and still be considered a treatment that has developed a substantial evidence base for its application.
# akha 30 Jan 2013 at 1:20 pm
This is Arif K. what is your scientific criterea?
[...] Science-Based Medicine » Is acupuncture as effective as … [...]
#jshiner A few reactions to your thoughtful post.
First, acupuncturists may well believe in their approach and the patients also, but belief is not enough to obtain results. The expectation that it is sufficient, is based on an understanding of results obtained in placebo conditions in clinical trials. These conditions do not simply reflect effects of belief, whether of practitioners or patients, but the sum total of background factors at work in a trial, including attention and support, ritual, or whatever.
Second, the larger acupuncture for depression literature has been shown in a number of reviews to be of exceedingly poor quality, with a strong confirmatory bias. Khan/Kirsch only picked out five studies that even their authors considered exploratory or preliminary. This is hardly the basis for the judgments that Khan/Kirsch made.
None of the studies selected by Khan/Kirsch involve head-to-head comparisons of acupuncture versus antidepressants or psychotherapy. They relied on voodoo statistics to make comparisons.
Psychotherapies on the other hand are manualized and the studies in which the therapies of validated allow therapists to improvise within the limits of these manuals.
Dr. Coyne,
Thanks for your reply. I want to clarify as well that my first paragraph in my prior post was in regard to Dirk’s postings.
I am trying to challenge your notion that 1) because you can think of proposed mechanisms for change that are assumed in evidence supported psychotherapies, allowing therapists to try a emphasize a different proposed mechanism of treatment benefit (ex. for CBT, switching from challenging distorted cognitions to behavioral activation), but 2) because you can’t think a proposed action for acupuncture, acupuncturists can’t make changes to treatment based on treatment response.
I am advancing that while assume effect of proposed mechanism of change after we see positive outcomes, good outcomes do not mean our proposed mechanism of change is accurate. Because of this, I offered that many acupuncturists can describe a proposed mechanism of how their treatment will be effective for patients, and use understanding of that mechamism/theory to engage with patients in responsive ways.
I am not denying that the studies on acupuncture for depression are not good quality and that there are no head-to-head studies with psychotherapies included. I am not denying this or advancing that acupuncture is an evidence supported treatment for depression. However, I think you left yourself weak in a argument in trying to make a distinction between the two treatments in that one has a mechanism for change and one doesn’t. I would propose that both can have active and (interally) consistent theories of change (regardless of the extent of helpfulness treatments impoart), and that the effectiveness of the treatments do not validate the proposed theory of change in themselves.
jshiner:
I believe what you are referring to is called “making $hit up” which is perfectly within the purview of acupuncturists since that is what they are doing to begin with.
I can see how an acupuncturist would certainly have more degrees of freedom in making things up to change therapy than a psychotherapist would.
I believe Dr. Coyne would say (well at least I would, and I think he would agree) that these changes in acupuncture aren’t really changing the actual therapy provided whereas in psychotherapy it is.
However, it is a valid point that the perception of change from the patient’s perspective, as told by the believing acupuncturist, could be indistinguishable from the valid evidence based changes of psychotherapy.
Whether this actually changes the outcome (i.e. would someone not responding to “traditional” acupuncture actually respond to such a “change” in the acupuncture regimen) is, to the best of my knowledge unanswerable with data at this time. I think it would not, on the average, change outcomes significantly (clinically at least, statistically possibly) but that is inference. A robust answer to that question would indeed by quite enlightening.
in my mind, dirk has moved from a blog commenter to a patient acting in an odd fashion that is a significant indicator of his or her mental difficulties. at this point, it jst doesn’t seem right to carry on the illusion of reasonable discussion and debate amongst interested, educated people.
dirk – print out this discussion, go to a decent therapist, and say, ‘here is part of my problem – i don’t have the ability to recognize it yet – please help me.’
jshiner: basically, here is some more info: the cog model of depression assumes that depressed people are depressed because of their beliefs/interpretations about the world arounf them. these are a certain type of beliefs that you could call ‘depressogenic.’
the theory says that a depressed person, with guidance from a counselor, can learn to identify these, can evaluate how they are unreaslistic, and can challenge these, and start thinking differently. if the depressed person starts perceving things in a less depressogenic way, they will start getting better.
there is more, but this is enough to answer the question.
decades ago, psychologists figured this out. And developed measures of this suspected phenomenon. Attributional Style Questionnaire is probably the leading one. we psychologists pretty much dominate the field of the science of measurement when it comes to difficult-to-measure phenomena – psychometrics. we have this one down pretty good. go read nunnallly. there is no scam or game. everything has to be on the up-and-up, replicable, etc.
If certain cogntive attributions are what is going on with depressed ppl, then you should see certain scores in depressed ppl, and different scores in non-depressed ppl. this is what they found.
also, as ppl get less depressed, the scores should change. this is what they found.
if a counselor strives to help a person change these thoughts, then depression should be relieved. this is what they found.
So, there is theory, and observed evidence that matches. and, the phenomena can be manipulated and predicted outcomes happen.
go google the ASQ.
In contrast, acupuncture posits that there is this life force flowing around in the body. it generally flows around in some certain way. illness results from it flowing in the wrong way somewhere or somehow.
Jshiner: you find me some reasonable, reliable measure of this life force.
you show me where clinical predictions match its quantity or value.
you show me where acupuncture is predicted to change it one way, a priori, then does.
show me replicability.
basically, you cannot.
at one point in time, in the 1970s, maybe the science of psychotherapy was somewhere close to the science of acupuncture. the practice surely was not that far off.
but we have been earnestly scientifically studying this since Eysenck criticized the practice of psychological counseling way back in the 1950s, pointing out we had no evidence of efficacy compared to doing nothing.
that kick in the pants got us rolling. by the 1970s, we had demonstrated efficacy of psychotherapy for depression. by the 1980s we were showing why/how.
the acupuncture people don’t have this. any of it. they have answers, but no measures, no studies, no supported predictions. they have nothing ruling out various likely biases such as expectancy effects and ‘non-specific therapeutic factors.’
that is what is different between acupuncture research and psychotherapy research.
there is no recognized
Dear bloggers: I would like to make a couple of statements that may help clarify the position taken by my colleagues and myself. First, in principle we agree that the methodology for assessing depression as well as designing and executing trials can be much better.
Having said, I would assert that the methods are a means and not the end. I would be hard pressed to accept that the data about the relative ease with which some depressives improve in a particular episode (up to 30%) is due to methodological flaws. In fact, focusing on methodological flaws lets you opt out of the real issue that over 50% of depressives don’t do well with the most relatively benign treatments, excluding treatments such as ECT, TMS, VNS or complicated pharmacotherapies.
Hence, the second point, that whatever we are dealing with is not entirely due to measurement problems. If we truly develop a ‘penicillin’ for depression, all this discussion about methods becomes mute, as the primafacie evidence would be simple enough.
So, focusing on the methods of measurement doesn’t lead to better therapeutic results. On the other hand, methods are simply a means to reveal false claims and to discourage shysters that sometimes disguise themselves as genuine advocates of help and hope for the depressed folks.
Thus, something more in the nature of understanding the disease and better treatments to be developed, while not being fooled by shysters, either by false criticisms and false claims.
Shysters, false criticisms, #Akhan? Can you point to any other meta-analyses and systematic reviews, other than those by your co-author Irving Kirsch, that rely on such voodoo statistics as you do in your review?
There are generally established standards for meta-analyses and systematic reviews. Why do you feel that you can flaunt them? Do you really believe you are inventing new methods of integrating data from diverse studies? Do you realize, that if your method of integration work, we would not even need randomized trials?
You are a true huckster, trying to pass off opinions as if they were based on a systematic review.
@medsvstherapy: I’d like to point out that the most effective treatment for depression is neither meds nor therapy. It’s ECT. I’m not dismissing psychotherapy at all, or medications for that matter. I’m just pointing out that the most effective treatment for depression does not line up with either of the two dominant theoretical frameworks for thinking about depression in the clinical world, and that is an indicator of how much progress still needs to be made.
@evilroboto:
Not sure how much you have seen or paid attention to my writings on the matter, but in a quick nutshell I have come to think of “depression” as a clinical syndrome that has at least 2 or 3 quite separate etiologies (neurotransmitter, neural pathway, and specific neural nuclei issues) that is a continuum and we simply do not have the technological or clinical sophistication to resolve the different etiologies, but the end result is a clinical picture we call “depression.”
Obviously a lot of nuance missing there, but I figured I would ask you specifically – does that seem like at least a sort of reasonable idea?
Sidebar:
“Depression” is too broad a term used too casually in too many contexts. Clearly, what rambly-troll-guy referred to as “depression” can be resolved through lifestyle changes. But what about those with a depression that doesn’t respond to treatment, pharmacological and otherwise?
Maybe because the brain is an organ and can malfunction just like any other organ in your body? Got type 1 diabetes? You don’t get better friends and a happier job so that your immune system decides to stop attacking beta cells in your pancreas; you take insulin injections. Got a mood disorder? Well, medical science is still figuring that one out.
@xplodyncow: it’s a very good point. The brain is the most complex organ in the body, so why would it make sense to believe that it’s somehow immune from dysfunction when other organs are not?
@nybgrus: I agree with you, and I would actually take it several steps further. It’s a slow moonlighting shift, so prepare for a rant.
Depression, even major depressive disorder as formally defined by the DSM, is a grouping of several distinct clinical syndromes that most likely has a large number of different etiologies, possibly even hundreds or thousands. And that’s not including bipolar depression, which often resembles the “atypical” subtype of depression but can also have mixed manic or hypomanic features, or the “depression”/dysphoria associated with borderline personality disorder (for example). The fact that these different subtypes of “depression” respond differently to different classes of medication indicates that they have different etiologies, but ultimately nobody knows.
Psychiatry is the only field left in medicine where all diagnosis and treatment is still based on clinical exam. This means that everything that psychiatrists treat is a syndrome rather than a clearly-defined disorder, with the exception of some types of mental retardation like Rett Syndrome that now have identified causes. I use the word “disorder” because that’s the terminology used in the field, but it’s more technically accurate to call them syndromes.
The only reason that it’s even possible for psychiatry to exist and maintain a surprisingly high degree of efficacy in the face of minimal knowledge of the actual pathophysiology is because different patients have such a high degree of similarity in their symptoms, which allows us to categorize people and find and validate treatments empirically. In my opinion, “depression” is the syndrome with the highest degree of clinical heterogeneity, more so than psychotic disorders, anxiety disorders, OCD (which falls between tic disorders and anxiety disorders), addiction, etc. I suppose you could say that “personality disorders” as a group are more heterogeneous.
Anyway, my opinion on the matter is that the major syndromes recognized by adult psychiatry (child psych is different) correspond to distinct failure modes of brain homeostasis. Depression is one type of broken state that the brain can fall into, just like addiction, anxiety, mania, psychosis, etc. We know how to induce states in humans, or at least animals, that are very similar to the primary disorders, which suggests that there is a surprisingly small number of ways the brain can break, and there are a large number of insults that can cause the brain to break down along one of those paths. For example, prolonged sleep deprivation causes depression in many people. If you lock a person up in a room long enough with a cocaine self-administration button, you could make them an addict (it works with rats). High doses of stimulants produce a state very similar to mania, including the psychosis that eventually results. This can be used as a model of psychosis itself, while psychosis plus the other two major symptom clusters of schizophrenia can be generated using NMDA antagonists like PCP.
At the end of the day, psychiatric disorders are just abnormal patterns of action potentials that the brain is unable to restore to a normal state, exactly analogous to cardiac dysrhythmias. And just like cardiac dysrhythmias, all we have to do to fix the problem is to restore the normal activity patterns, or at least a less abnormal activity pattern that is grossly compatible with normal physiological function. In the heart, there are dysrhythmias that we can’t fix, but we can lesion the conduction pathways and pace the ventricles, and the patient’s fine. In the brain, it’s more complicated. However, in either the heart or the brain, if fixing the activity patterns required us to actually understand the underlying pathophysiology fully or to manipulate each cell individually to restore normal activity patterns manually, it would be completely hopeless. Fortunately, we don’t actually have to do those things most of the time – all we have to do is bump things in the right direction enough to allow brain (or heart) homeostasis to fix things. That’s why psychiatrists can prescribe drugs with incredibly broad action across multiple parts of the nervous system and still get therapeutic effects. It’s not because psychiatric disorders have anything to do with a primary dysfunction of neurotransmitter signaling, i.e. a “chemical imbalance” or anything like that. It’s that creating artificial chemical imbalances can actually induce or relieve the symptoms by changing the firing patterns.
Thanks for that response evilroboto. I don’t know if you actually have been reading what I tend to write on the topic, but that is very, very much in line with the thoughts I have been forming on the topic myself. Not nearly as refined as yours, quite obviously, but that is the gestalt of what I have been inferring from my own readings of the evidence and literature on the topic. I feel like I deserve to give myself a little pat on the back now, so thanks!
And of course xplodynowcow – very succinct and also quite in line with my own thoughts, so thanks to you as well. I’ll be using that analogy from now on though I hope you’ll understand if I don’t credit you every time
This is a perspective I hadn’t actively thought of and actually helps me quite a bit, so thank you again. I had an argument with someone online many moons ago about it and had I thought of this perspective I think we may not have argued as much. However, it will help me with a new frame through which to look at future learnings on the topic (not my only frame, but merely another way to look at things as I synthesize my ongoing analyses and thoughts as I continue my learning).
I am not interested in pursuin a career in psychiatry, but I do find the field interesting, overly and unreasonably maligned, and genuinely an area of medicine with significant promise for expansions and becoming more robust as our technological sophistication grows along with our corpus of knowledge.
Thanks again for taking the time to post!
@nybgrus: sure, I’m glad you found something about what I said interesting.
While responding to a pro-acupuncture commenter on my blog about the lack of positive double-blind trials for acupuncture, I came across this recent study on PubMed (Feb/2013)
http://www.ncbi.nlm.nih.gov/pubmed/22943846
“Acupuncture improves sleep in postmenopause in a randomized, double-blind, placebo-controlled study.”
Conclusion: “Acupuncture was effective in improving reported sleep quality and quality of life in postmenopausal women with insomnia.”
Judging from the abstract, it seems to me that the conclusion is not reasonable. They found an improvement in ONE metric, in a small study of 18 patients overall.
And also they say:
“Comparison of baseline and post-treatment data of the acupuncture group showed that treatment resulted in significantly lower scores on the Pittsburgh Questionnaire and an improvement in psychological WHOQOL.”
Meaning the test subjects “improved” after treatment – that’s really astounding. And no word on comparing the improvement to the control group.
*****************************************
Posted the above to topic suggestions on Neurologica, and thought I might post it to a recent SBM text on acupuncture as well.
that’s the rub Nikola. You can call something a double blind RCT and it is still worthless. Or at least nearly so. Assuming the study actually had adequate blinding (which is always a reasonable question in acupuncture studies and especially in such small studies) the study size is just way to small to say anything at all. And it sounds like the went for multiple endpoints which makes it way too small. It is essentially a grown-up version of a high school science fair project.
Thanks. That’s what I imagined, however, since I don’t have any formal scientific training I like to get a second opinion about my interpretations of studies. Unlike the “scientists” who performed that study, I might add.
I’ll do a rapid run down from the very beginning.
1) Climacteric (the journal) has only been around since 1998. Not necessarily a bad thing but could mean it is a shady journal.
2) Impact factor 1.98. Also not necessarily bad, could bu just because it is new. Or because nobody cites it for a good reason.
3) The selection criteria basically was for perfectly healthy post-menopausal women aged 50-67
4) groups were acupuncture and sham acupuncture (which they call “placebo”)
5) Twice a week for 5 weeks
6) They say ther were double blinded but they didn’t say how and their phrasing was odd:
“Neither the researchers nor the participants knew which patients had received the sham acupuncture and which had received the actual acupuncture, to prevent bias on the part of the researchers and the volunteers.”
I mean it is correct, but I don’t recall reading a paper that told me why we double blind. I know why we double blind, you know what I mean? Just struck me as odd is all.
7) They started with a pool of 102 but only 18 met DSM-IV insomnia criteria so that was the study size.
8) Once again some wierd phrasing:
“At baseline, the two groups were not significantly different in anthropometric characteristics such as BMI and age. There were no differences between the variables of age and BMI when comparing the two groups, showing comparative viability between the acupuncture”
9) It seems that they demonstrated improvement in each group from baseline but no difference between the groups. In other words they intervention they themselves call “placebo” worked in exactly the same way as the actual acupuncture (I know, shocking right?).
“The acupuncture group showed a significant improvement in the PSQI questionnaire when baseline and final evaluation were compared (12.77 ± 0.7 vs. 9.77 ± 0.8, respectively; p < 0.01; effect size = 0.8) and an enhancement of the psychological WHOQOL (57.07 ± 4.1 vs. 66.67 ± 3.1, respectively; p = 0.03; effect size = 0.66)"
"The comparison between groups after treatment showed no differences among questionnaires, but the comparison did show a trend toward improvement in PSQI in the acupuncture group (9.77 ± 0.8 vs. 12 ± 0.9; p = 0.058; effect size = 0.45).”
And also it is funny that they mention the psychological WHOQOL as getting better. I wonder why they didn’t mention that they were looking at 6 comparisons and only that one showed improvement, and that in the sham group the social WHOQOL improved as did the environmental WHOQOL?
So in other words, no difference between “treatment” and “placebo” but somehow the placebo actually had more positive effect than the treatment!. 2/6 vs 1/6 metrics improved. The “placebo” also had improvements in sleep latency. They had another 8 measurements they were looking at, of which “treatment” improved on one, maybe two.
So we have 14 endpoints they are separately analyzing on a group of 18 people, with both placebo and treatment showing improvement over time, but no difference between the two groups, and in fact placebo outperforming (a bit) the treatment group.
Sounds like a resounding endorsement of acupuncture to me! After all, in the words of the authors:
“Acupuncture intervention produced a significant improvement in subjective sleep quality, as evaluated by the psychological WHOQOL and sleep quality indicated by the PSQI at the final evaluation. This finding suggests that the acupuncture treatment was effective for treating insomnia, as suggested in other studies that showed an improvement in sleep quality through acupuncture by different types of evaluations, such as melatonin secretion and nocturnal hot flushes”
Clearly that is what the study showed, after all.
They even say that their study is just like others that have shown:
“A systematic review of randomized, controlled trials using meta-analyses to study acupuncture for the treatment of insomnia31 showed a beneficial effect of acupuncture compared to no treatment and of real acupressure compared with sham acupressure on total PSQI scores.”
Wait, I thought we were talking about acupuncture, not pressure. Oh well, must be the same thing. And yes, clearly this study showed us that “real” vs “sham” acupuncture was so much better.
I’ll let the authors have the last words:
“Based on the results of this randomized, controlled-trial study, we conclude that acupuncture is effective in improving the quality of sleep and the psychological domain of quality of life in postmenopausal women with insomnia. Furthermore, we observed an important decrease in subjective insomnia supported by a high effect size.”
[...] There is never a shortage of people claiming that antidepressants are a Big Pharma Conspiracy, that MDD is a myth or can be cured by diet or supplements or exercise or thinking happy thoughts. The problem is that these claims are based upon false information or a lack of understanding or, sometimes, willful ignorance. James Coyne addresses the problems with the reports that antidepressants are no more effective than placebo here and here. [...]
Thanks for the analysis, nybgrus. Excellent work!