Ghosts, UFOs, Yeti and Antidepressants? February 26
A new study1 suggesting that the benefits of SSRIs (antidepressants like prozac) don’t provide clinically significant benefit to most patients has been making the rounds in the news today. Relying on unpublished studies submitted to FDA during the approval process for various SSRI drugs the authors did a meta-analysis and only found a clinically significant difference between the drugs and placebo for the most severely depressed patients, and only then because these patients seem to respond less to placebo not because they respond more to the medication. The news is reporting this as if this class of antidepressants (still considered some of the most effective medications) only makes a big difference for the very depressed but for once the news may be understating the scientific evidence. The truth is that it’s not clear if antidepressants work at all.
The problem is that the trials used to test antidepressant efficacy compare the medication to a sugar pill but most patients and their doctors are able to tell when they aren’t on a sugar pill. Given the small benefits of these antidepressants over placebos the fact that study participants are able to break the blind may well explain the entire effect. For those who are interested I’ll lay out a full argument for this position below the break. However, it’s important to note that just because antidepressants might not work isn’t any reason to stop prescribing them. So long as they give greater response than placebos (i.e. people believe in them) they may be of use to some people. Alternatively one might think these drugs really do have a *very *minor effect in the most depressed individuals, if for no other reason than a system that is sufficiently far from the mean might be (on average) pushed closer by any substantial intervention (think wacking your ipod to make it work) amplified by the placebo effect.
These considerations add weight to the view that depression is not a disease or even condition in the way that Downs syndrom or Asthma is but is instead the extreme end of a distribution. Thus rather than being the result of a determinate dysfunction as those cute Zoloft commercials would have you believe having depression is more like being short2. This theory is being advanced by some serious researchers3 4. For some reason, however, these attitudes seem to frequently come part and parcel with the idea that we ought not to treat ‘natural’ sadness, in this case apparently justified by the evolutionary ‘argument’ that sadness must serve some purpose (of course an inclination toward or efficacy at rape might have been selected for as well it would hardly prove it was something important to preserve in modern society). While some would dispute this theory everyone, it seems, is obsessed with distinguishing normal and abnormal sadness.
I could not disagree more forcefully. Far from being an argument to abandon the idea of alleviating suffering via chemical intervention the idea that depression is just an extreme for of unhappiness, if true, militates for research into effective drugs to elevate mood. Would we deny those so short they experience serious hardship or those with GH deficiencies in adulthood treatment with growth hormone just because some of it’s benefits are available to those with average levels of the hormone? Of course one always has to balance benefits with side effects and depending on the drug that balance may very well weigh against treatment in those who are already quite happy (as I understand does for HGH treatment for most people) but it’s hard to imagine that any bearable level of side effects is enough to justify denying an effective treatment to those who are chronically suicidally depressed.
Placebo Response and Theory Failure
Even before this recent meta-analysis the available data indicated that placebo response was at least 80% as effective as administering the antidepressants themselves5 and the new data only pushes the effects down below the level of clinical significance (i.e. the level generally considered worthwhile to prescribe). Generally in the more recent studies the percent of patients who ‘respond’ to the active medication is centered near 50% while those responding to placebo is around 30% but the response rate from placebos is highly variable (10%-50%) and interestingly increasing over time6. Even those who defend antidepressants acknowledge that remission rates only differ by 20% (about 50% on placebo versus 70% on medication) and merely argue that this indicates a real result that can be extremely significant in the ‘true responders’7.
Were these improvements the result of a reasonable theoretical hypothesis about the cause of depression, or at least the mechanism of action of antidepressants we might have more reason to believe this weak evidence indicated something real. However, not only do we not really have a good theory for how SSRIs work (no immediate effects are observed but seretonin levels quickly drop back to normal invalidating the initial theories) but there are similar response rates from a diverse array of ‘antidepressant drugs’ with no apparent common mechanisms nor ability to tell which patients will respond best to which drugs8. At the very least this sort of situation should raise concern about whether there is any real effect and if so whether it is primarily due to some kind of regression to the mean rather than specific efficacy against depression.
Active Placebos and Unblinding
The most daming criticism about the evidence for antidepressants is the concern that the supposed effects of the drug come entirely from unblinding9. There is evidence that patients are able to figure out when they are taking the real drug versus a sugar pill10 11. Moreover, other studies suggest that active placebos are more effective in treating depression than inactive ones12. In fact it was the defense offered for antidepressants that many of these ‘active placebos’ have antidepressant activity themselves that was particularly convincing to me that there might be some element of bias in the evaluation13.
Ultimately I can’t be sure which is ultimately the right answer. These arguments certain raise troubling questions about the effectiveness of antidepressants (and there is plenty more I just ran out of steam). Additionally the profusion of backup theories offered to save the hypothesis that these are antidepressant drugs seems dubious. Doctors tell patients they just need to find the right drug, researchers claim that it’s really just that a few ‘true responders’ are hiding in the data or that the ‘active placebos’ themselves have antidepressant effect. All of this on top of any publication bias or other bias induced by the strong motivation many people have to believe these drugs are effective.
-
Kirsch, I., Deacon, B. J., Huedo-Medina, T. B., Scoboria, A., Moore, T. J., and Johnson, B. T. (2008). Initial Severity and Antidepressant Benefits: A Meta-Analysis of Data Submitted to the Food and Drug Administration, PLoS Medicine, 5(2), e45 ↩
-
Of course like being exceptionally short we would expect some people with exceptional depressions to have it as the result of some abnormality but not necessarily all the same one. The important point is that many of the people who have even severe depression may lie on a continuum with those who are just slightly less sad than normal and even those who are more happy than average. ↩
-
Moncrieff, J. (2007). Rebuttal: Depression Is Not a Brain Disease, CANADIAN JOURNAL OF PSYCHIATRY, Vol. 52 ↩
-
Horwitz, A. V. and Wakefield, J. C. (2007). The Loss of Sadness: How psychiatry transformed normal sorrow into depressive disorder ISBN: 9780195313048 ↩
-
Kirsch, I., Moore, T. J., Scoboria, A., and Nicholls, S. S. (2002). The Emperor’s New Drugs: An analysis of antidepressant medication data submitted to the US Food and Drug Administration, Prevention & Treatment, 5(1), ↩
-
Walsh, B. T., Seidman, S. N., Sysko, R., and Gould, M. (2002). Placebo Response in Studies of Major Depression: Variable, Substantial, and Growing, JAMA, 287, 1840-1847 ↩
-
Michael E. Thase, M. D. (2002). Small Effects Are Not Trivial From a Public Health Perspective, Psychiatric Times, 19(9), ↩
-
Hindmarch, I. (2002). Beyond the Monoamine Hypothesis: mechanisms, molecules and methods, European Psychiatry, 17(Supplement 3), 294–299 ↩
-
Kirsch, I. and Sapirstein, G. (1998). Listening to Prozac but hearing placebo: A meta-analysis of antidepressant medication, Prevention & Treatment, 1(0002), ↩
-
Moncrieff, J. (2001). Are antidepressants overrated? A review of methodological problems in antidepressant trials, J Nerv Ment Dis, 189(5), 288–295 ↩
-
Piasecki, M. P., Antonuccio, D. O., Steinagel, G. M., Kohlenberg, B. S., and Kapadar, K. (2002). Penetrating the blind in a study of an SSRI, Journal of Behavior Therapy and Experimental Psychiatry, 33(2), 67–71 ↩
-
Moncrieff, J., Wessely, S., and Hardy, R. (2004). Active placebos versus antidepressants for depression., Cochrane Database Syst Rev, (1), CD003012 ↩
-
Salamone, J. D. (2000). A critique of recent studies on placebo effects of antidepressants: importance of research on active placebos, Psychopharmacology, 152(1), 1–6 ↩
Loading ...
[...] tagged placeboOwn a Wordpress blog? Make monetization easier with the WP Affiliate Pro plugin. Ghosts, UFOs, Yeti and Antidepressants? saved by 5 others wineypooh bookmarked on 02/27/08 | [...]
[...] TruePath always has something good to say. I like this one posted earlier today. Follow the link for the whole thing.The problem is that the trials used to test antidepressant efficacy compare the medication to a sugar pill but most patients and their doctors are able to tell when they aren’t on a sugar pill. Given the small benefits of these … [...]
All of us have different DNA and different metabolisms. Some herbs and vitamins work better than others. The first thing that is needed is proper nutrition and a good physical exam. As the director of Novus Medical Detox, I often see patients who are on alcohol or opioids, central nervous system depressants, also taking antidepressants. When they detox they find they don’t need the antidepressants.
This is good news because a Swedish study showed that 52% of the 2006 suicides by women on antidepressants. Since antidepressants work no better than placebos and are less effective than exercise in dealing with depression.
There is a prescription drug epidemic and these are leaders in the list of terribe abuses.
Steve Hayes
http://novusdetox.com
I’m not disputing the idea that some drugs and vitamins almost certainly have somewhat different effects in different people (though how significant these differences are or in what if any areas we know enough to exploit them is something I don’t know). Indeed if we had some kinda of effective means of predicting which antidepressants would be best for which people that would be strong evidence for their effectiveness.
However, the problem I am alluding to is this. Studies of depression have long found that over periods of several months (the amount of time people often try out a new antidepressant) about 1/3 of sufferers get worse, a 1/3 get better and 1/3 stay the same (very roughly). Thus it should make us suspicious of the efficacy of these drugs if practioners tell us (contrary to what we would expect from most theoretical models of how these drugs work) that they have to try many of the antidepressants until they ‘hit’ upon the right one for the patient as this is exactly the observed results we would expect from spontaneous remission plus placebo effects. In other words the confidence many practitioners have in thee drugs at first glance might seem compelling until we notice that it look suspiciously like the sort of inference humans are liable to draw in an attempt to explain what are largely spontaneous or poorly understood events. Especially given that similar claims were made for psychoanalysis until they were ultimately debunked (amusingly this was one of Timothy Leary’s big contributions to psychology).
As far as your remarks about alcohol and opiates it’s certainly true that the abuse of these substances can accentuate and worsen depression. However, one has to be careful about the degree of causality to assign to theses agents given the fact that abuse is often something people do in reaction to their underlying unhappiness. Unsurprisingly those who feel like they have nothing to lose and are in great pain are more likely to turn to various drugs. Conversely successful recovery from drug abuse will (from what I’ve seen) often coincide with some improvement in their underlying psychological problem (whether as the result of intervention by drug abuse professionals, self-reflection or simple time). Certainly none of this denies that the abuse plays a role in worsening the situation but separating out the magnitude of the various effects is very hard.
Also it’s important not to confuse the effects of abusive use, with it’s attendant large dosage fluctuations, from medically supervised use. Despite the bad rap that opiates gets those who are given them for postop pain have very very low rates of resultant addiction and studies of medically controlled uses of either weak (codeine) or partial agonists have suggested anti-depressant activity (though one should be skeptical just as I am here). Just because the major opiates are abused to make depression worse doesn’t mean we should rule out opiate derivatives as potential anti-depressants to replace the ineffective ones on the market today.
Whatever might be true about how many people can be helped there are certainly a clear minority who are afflicted with a long term, treatment resistant severe depression. If these people had terminal cancer or AIDS we would be clamoring for treatments, even experimental treatments that imposed other serious risks. When someone is so depressed their life just isn’t worth living and doesn’t improve for many years we should be making risky dangerous interventions like seeing if prescriptions to opiate compounds or adderall or what not will improve their life even though these would be unthinkable risks for the vaguely depressed normal patient.