tag:blogger.com,1999:blog-4477385342066275897.post329741217715139690..comments2024-01-11T10:42:04.473+00:00Comments on Hawk/Handsaw: More meta-analysis delightPaul Wilsonhttp://www.blogger.com/profile/18101626906004768474noreply@blogger.comBlogger9125tag:blogger.com,1999:blog-4477385342066275897.post-81829112216764970362008-10-27T20:27:00.000+00:002008-10-27T20:27:00.000+00:00Thanks. You are now my first link...Thanks. You are now my first link...Neuroskeptichttps://www.blogger.com/profile/06647064768789308157noreply@blogger.comtag:blogger.com,1999:blog-4477385342066275897.post-45015623915364522762008-10-27T14:17:00.000+00:002008-10-27T14:17:00.000+00:00Neuroskeptic: cheers for that, and thanks for the ...Neuroskeptic: cheers for that, and thanks for the link to your blog. Excellent first post. I'll bung up a permanent link in the sidebar.Paul Wilsonhttps://www.blogger.com/profile/18101626906004768474noreply@blogger.comtag:blogger.com,1999:blog-4477385342066275897.post-82458946494662198272008-10-26T14:17:00.000+00:002008-10-26T14:17:00.000+00:00That's really very silly. Although not as silly as...That's really very silly. Although not as silly as believing in homeopathy in the first place.<BR/><BR/>Speaking of post-hoc data dredging, Lynn McTaggart of Intention Experiment fame (and friend of homeopathy) has recently surpassed herself in this arena. She's managed to take data showing that her latest Intention Experiment killed dozens of people, and claims that it's a success.<BR/><BR/>I hereby shamelessly plug my new blog:<BR/><BR/><A HREF="http://neuroskeptic.blogspot.com/" REL="nofollow">Neuroskeptic</A><BR/><BR/>where I have written about this... it makes her previous experiments with Rustram Roy look like Nature material.Neuroskeptichttps://www.blogger.com/profile/06647064768789308157noreply@blogger.comtag:blogger.com,1999:blog-4477385342066275897.post-75375701737064333232008-10-26T14:15:00.000+00:002008-10-26T14:15:00.000+00:00This comment has been removed by the author.Neuroskeptichttps://www.blogger.com/profile/06647064768789308157noreply@blogger.comtag:blogger.com,1999:blog-4477385342066275897.post-91831514171389091692008-10-22T12:11:00.000+01:002008-10-22T12:11:00.000+01:00jdc wrote, "Didn't Linde et al ... later conclude ...jdc wrote, "Didn't Linde et al ... later conclude in another paper that the conclusions of their meta analysis were weakened by their more recent work that looked at trial quality?"<BR/><BR/>Yes. Not only did they conclude that the evidence of bias they detected by their reanalysis of the same data "weakens the findings of our original meta-analysis", but also that an update to their analysis of studies of classical homoeopathy, and the fact that a number of new high quality trials (not included in their analysis) had negative results "seem to confirm the finding that more rigorous trials have less-promising results." They concluded from all this that their 1997 paper "at least overestimated the effects of homeopathic treatments.<BR/><BR/>Linde K, Scholz M, Ramirez G, Clausius N, Melchart D, and Jonas W. <I>Impact of Study Quality on Outcome in Placebo-Controlled Trials of Homeopathy</I>. J. Clin. Epidemiol. 52 (7) 631-636 (1999)<BR/><BR/>Linde and Jonas also commented, in the course of a letter to the <I>Lancet</I> critiquing the Shang paper, that "our 1997 meta-analysis has unfortunately been misused by homoeopaths as evidence that their therapy is proven."<BR/><BR/>Not that this has had any effect on the homoeopaths, of course.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-4477385342066275897.post-17036565429480108832008-10-22T11:27:00.000+01:002008-10-22T11:27:00.000+01:00pj:To be fair, Shang is perhaps a little confusing...pj:<BR/><BR/>To be fair, Shang is perhaps a little confusing on this point. Although they talk about "larger" trials throughout the paper, they're actually looking at "lower SE" trials. Still, if you're going to criticise the way they picked their "larger" trials, you should probably read the paper to find out how they did it first...Paul Wilsonhttps://www.blogger.com/profile/18101626906004768474noreply@blogger.comtag:blogger.com,1999:blog-4477385342066275897.post-30879234179528003942008-10-22T11:23:00.000+01:002008-10-22T11:23:00.000+01:00jdc:Yes, that's a good point. Homeopaths seem obse...jdc:<BR/><BR/>Yes, that's a good point. Homeopaths seem obsessed with the idea that Shang was somehow biased and/or fraudulent. This prevents them from actually looking at the paper objectively. In fact, it's deeply unsurprising that they should have come to different conclusions to those of Linde et al., for the reasons you point out.Paul Wilsonhttps://www.blogger.com/profile/18101626906004768474noreply@blogger.comtag:blogger.com,1999:blog-4477385342066275897.post-23586544808821448802008-10-21T17:41:00.000+01:002008-10-21T17:41:00.000+01:00Out of interest - in the 'background' section they...Out of interest - in the 'background' section they refer to "a discrepancy between the outcome of a meta-analysis published in 1997 of 89 trials of homeopathy by Linde et al and an analysis of 110 trials by Shang et al published in 2005". Didn't Linde et al though: (a) find no evidence that hpathy worked for any particular condition; and (b) later conclude in another paper that the conclusions of their meta analysis were weakened by their more recent work that looked at trial quality?<BR/><BR/>If Rutten and Stolper were worried about this discrepancy and this is their main reason for publishing in Homeopathy, could it be the case that their efforts are in vain? It seems to me that Linde et al may have already provided a convincing explanation of this discrepancy.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-4477385342066275897.post-28205349935133554102008-10-21T16:57:00.000+01:002008-10-21T16:57:00.000+01:00Presumably they are also unaware of the concept of...Presumably they are also unaware of the concept of standard error, hence their comments concentrating on sample size.pjhttps://www.blogger.com/profile/06832177812057826894noreply@blogger.com