Posted by Cam W. on October 29, 2000, at 10:55:56
In reply to Re: Cam - Peer reviewed vs non peer? Reliable?, posted by JohnL on October 29, 2000, at 5:22:15
John -
It makes a HUGE difference as to who is paying for the study. Because of this being such a litigous society, most clinical trials are like Wonder Bread - homogenous, but lacking substance.
Clinical trials contain results from right-handed white males, 25 - 50 years, who have no comorbid disorders, do not drink to excess or do drugs, and who can be talked into continuing to take a drug, no matter how bad the side effects are.A guy that I know socially was in a drug trial for a new heart drug. He came to me and said that this drug he was taking was causing him to act "goofy", gave him headaches, and made him pee his bed every night. He had talked to his doc about this and the doc said the trial only last 8 more weeks (12 week trial) and to put up with the side effects, in the name of science. It is one of the only times I have told someone that maybe they should find another doctor (there was also more involved, including the doc being paid for each patient he could keep in the trial).
As for randomized, double-blind, placebo-controlled trials; do they really exist? It has been found that blinded researchers and even saavy guinea pigs (..er, test subjects) can tell, with a greater liklihood than chance, if they were taking a placebo or the drug. This can be suggested by the side effects profile or the action of the drug (eg How would you design a double-blind, placebo-controlled study of LSD?). Using an active placebo (eg like atropine in LSD studies) can help, but are not perfect. Using a placebo and an active comparator (eg a TCA in a SSRI study) can also help but is not perfect, either. I know one guy who was in a Seroquel study and was given Haldol. He knew this because he had tried all the different antipsychotics several times. He knew how each "felt". He even told me he was taking Haldol. I looked into it and after he had dropped out and the seal was broken, indeed, he was taking Haldol. Perhaps this treatment resistant guy would be a great teat subject (non-compliant, abuses a variety of street drugs, has very developed delusional systems, will go out of his way to help you, and is one of the most honest, reliable, down-to-earth people that I know). This gentleman (who is truly a gentle man) needs to receive his meds daily because he cannot trust himself not to take a week's worth of meds in 1 or 2 day and is floridly psychotic 24/7. Help HIM, drug companies!
Perhaps you can't skew clinical results, but you can fudge as to what they are saying. For example, I saw the following in an article about mirtazapine: "it has the added advantage over other antidepressants of increasing appetite - especially good in wasting diseases - and causing sedation - especially good for insomnia". Also, studies do not seem to be funded more than 6 weeks in many cases. Hell, most drugs don't show full effect that quickly (eg OCD cases). So, if you look at the product monograph of Effexor you will see that the top 5 side effects (nausea, headache, somnolence, dry mouth and dizziness) are all start-up side effects and usually disappear within a couple of weeks. Sexual dysfunction didn't crack the top ten, nor did weight gain; two side effects that do often appear after 2 or 3 months of taking the drug. The PDR (Physician's Desk Reference - US) and the CPS (Compendium of Pharmaceuticals and Specialties - CDN) are filled with drug monographs that read like lawyers' documents, written by lawyers for lawyers.
What a clinical trial can show is gross problems that can happen when taking the drug (eg heart problems in many taking sertindole) or it's relative effectiveness versus other similar clinically tested drugs. Naturalistic studies have to be done, including tens of thousands of people with the disorder. Many of these "real world" people smoke pot or crack, drink to excess on occasion, are female, have congestive heart failure (or other comorbid disorder). Let's see how these drugs do in the real world, head-to-head against it's competition.
Early responders to drugs which usually have a delayed actions in most other people is somewhat of a quandry. Many are placebo effect, a result of actually doing something about your illness or problems, but not all. In many other cases early response is mistaken for secondary side effects (anticholinergic and antihistaminergic side effects of TCAs). These secondary effects could be enhanced to look like drug effects, depeneding upon that person's brain's chenical mix (eg giving an SSRI to someone with noradrenergic overload may decrease some of the NE caused effects, thus the person seems to be responding to the drug. This could change in a week or two as the SSRI begins to change that internal mix). You have to wait with any psychoactive drug, because the body is going to try to overcome the effects of the drug by changing concentrations of neurotransmitters and their receptors in different parts of the brain. Every person's internal environment is different, but when a vast majority of people have a delayed response in treating a disorder, you look to other causes for improvement (other than drug effect).
As for your physicians as liars comment: why should they be any different from you and me? These researchers are trying to pay mortgages, "keep up with the Jones", and carve a niche in the scientific framework with their name etched on it. Most negative studies that are supported by the drug company aren't reported and the researcher isn't the first asked to do the next trial. Results from trials are in many cases subjective. The concrete results can also be manipulated. For example, what is the criteria for response (usually a 50% reduction in HAM-D scores for AD testing). This 50% reduction in HAM-D scores doesn't say how many of the residual depressive symptoms are remaining. Someone who is severely depressed at onset of therapy may still be significantly depressed with a 50% reduction in their symptoms.
Researchers are becoming very deft at reporting what both the drug company and the research journal want to hear. Look at the fine print of the journal article, not only for who's paying the bills, but also how much trouble it was getting the article printed. Look for the following dates in the fine print. Date received, date accepted (after changes made) and date published.Some articles aren't published until over a year after acceptance. This, in my opinion, reduces the worth and applicability of the study. It shouldn't, but it does. I figure that if Dr.Stahl (J Clin Psych) or Dr.Andreasen (Am J Psych) added this study "to fill space", a year after it was accepted, this study must not be as important as others in the same journal.
You actually posed these questions to me at a time when I have been reading and highlighting about 10 journal articles a day. I am quickly becoming jaded to not only the whole drug approval process, but also the elegantly written drug commercials in the form of journal articles.
Anyway, before I get myself in trouble with the drug companies and never get another speaking engagement (yes, I'm a drug company whore, but I try to minimize my worship of any drug), I think I will stop here.
Scientific method isn't perfect, but until someone can show us a better way of proving things to be objectively true (sorry, dreaming again), scientific method is all we have. People will always find ways to make their product look better than their competitor's (eg Vick's inhalers make you "feel" clearer, they don't make you clearer).
1) If it sounds to good to be true, it is.
2) Believe nothing of what you hear and only some of what you see.
3) Never believe that someone has "your" best interest in mind, particularily if it involves the output of money.One of my longer rants - Cam
poster:Cam W.
thread:47596
URL: http://www.dr-bob.org/babble/20001022/msgs/47697.html