 |
 |
i got into another little exchange with dr. doraiswamy (co-author with dr. szarfman on study #3) between naps while i was home sick this week, and had an epiphany. those two studies - the two gambling/DA stories - they actually show that there is *not* an association between DAs and gambling. so far, both a friend of mine and dr stacy have confirmed that as far as they, neither of whom is a statistician, can tell, my calculation of significance is correct - in other words, there is none - so that study actually shows the opposite of what it was publicized as showing. and dodd's 2005 study - i can think of no reason that an author would withhold the number that would allow the determining of incidence - except one - that number would have shown that there was no association between DAs and gambling. and *that* means that the FDA "study," which they unabashedly announce in the opening salvo is meant to expand on dodd's 2005 study, given that most, if not all, the reports it finds came in after the 2005 study, that means the FDA is reporting on a signal that is based not just on reports that came in as a result of publicity of a phenomenon with no basis in reality, but on reports that came in as a result of publicity of a phenomenen that had actually been shown *not* to exist. i hope that makes sense. m mackenzie <[log in to unmask]> wrote: mackenzie wrote: The third study, published in February of 2006, was actually a sort of elevated letter to the editor (LTTE) sent to the Archives of Neurology (AN) in response to Dr. Dodd�s study. The lead author, Dr. Ana Szarfman, works at the Center for Drug Evaluation and Research (CDER) at the FDA.* Dr. Szarfman and colleagues expanded on Dr. Dodd�s study by �mining� the FDA�s Adverse Event Reporting System Database for reports of gambling associated with particular drugs. They found 67 reports associated with 23 drugs* - Mirapex had 39 of those reports. They conclude their report by saying that given their findings in the database, and the reports of the onset/cessation of behavior coinciding with commencing/discontinuing �these� drugs, and the case reports in the literature, the question should be raised ofo whether these drugs have an effect on impulse control. I have more issues with this study than you could shake a stick at, but i will only address the two that i see as most important. First of all, there is the caveat found in the instructions regarding how the database should and should not be used - i think it speaks for itself: "For any given report, there is no certainty that a suspected drug cause the reaction. This is because physicians are encouraged to report suspected reactions; however, the event may have been related to the underlying disease being treated, or caused by some other drug being taken concurrently, or simply occurred by chance at that time. Accumulated reports cannot be used to calculate incidence (occurrence rates) or to estimate drug risk. Comparisons between drugs cannot be made from these data." Secondly, given that the 2003 and 2005 studies are arguably fatally flawed, the only way a report of gambling with Mirapex could be credible is if it was registered before the 2003 study was published. So, I asked Drs. Szarfman and Doraiswamy to tell me the dates of the reports, or at least how many came in before August on 2003 and how many after. I asked for that information three times. The first time, Dr. Szarfman sent me to a place called NTIS, where I discovered that it would cost me about $4000 to obtain the whole database back to 1968, and that I would need some kind of industrial strength database program on my computer into which the data could be dumped. I went back to Dr. Doraiswamy asked again, twice. By the time I made the third request, I had become quite sure that the previous two studies were baseless and so I asked Dr. Doraiswamy to address the two biggest holes in those two studies - the question of statistical significance and the question of a lack of incidence - and then I requested to be given the dates once more. Dr. Doraiswamy responded wishing me good luck on my research and saying that he did not have the information i sought handy. I didn�t respond until today at which time I wrote and asked for some clarity on an apparent discrepancy in the study (see asterisk below.) I have submitted a request under the Freedom of Information Act to pull all the information I think Szarfman et al pulled for their study.. I expect to have it in about two months. While writing this I have realized there is a Part IV. sigh. And maybe part V. Mackenzie *The 67 reports distributed amont 23 drugs part appears to be contradicted by a table showing 67 reports apparently distributed among 6 drugs - i have another email in to Dr. Szarfman��s colleague, Dr. Doraiswamy. __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com ---------------------------------------------------------------------- To sign-off Parkinsn send a message to: mailto:[log in to unmask] In the body of the message put: signoff parkinsn --------------------------------- ---------------------------------------------------------------------- To sign-off Parkinsn send a message to: mailto:[log in to unmask] In the body of the message put: signoff parkinsn Yahoo! Messenger with Voice. PC-to-Phone calls for ridiculously low rates. --------------------------------- ---------------------------------------------------------------------- To sign-off Parkinsn send a message to: mailto:[log in to unmask] In the body of the message put: signoff parkinsn Blab-away for as little as 1¢/min. Make PC-to-Phone Calls using Yahoo! Messenger with Voice.