 |
 | Home | Contact Us | FAQ | Search & Site Map | Link to Us |
 | Sign In | Join | Other 45 Sites in Network |
Medical Forum / Diseases and Disorders / Tinnitus / April 2005How to report the result from a medical study?
The result from a study, be it double or triple blind, controlled or otherwise, should be expressed in the format of 'probabilities, degree of response, chances of success, etc.' and to do so in conjunction with the underlying conditions surrounding a particular ill. The result should not be expressed in the dogmatic format of a simple dichotomy between whether 'it works or does not work'. The latter is misleading to patients. The reason is: the result of a study is only a result, and this result does not _necessarily_ represent 'truth'. Since it is not _necessarily_ truth, it should not take the form of the absolute...IMO. FP > The result from a study, be it double or triple blind, controlled or > otherwise, should be expressed in the format of 'probabilities, degree [quoted text clipped - 7 lines] > > FP This isn't applicable to the results of any double blind study I've reviewed. Every one I've seen publishes percentages of people in each category (active/placebo) who report a degree of relief from whatever ailment the treatment is tested against. The underlying condition may or may not be treated. For example, if one is testing the effect of a magnetic field on healing a broken bone, the underlying condition might be carelessness and/or stupidity if the bone was broken in an accident while driving drunk. "Bill" <Bill@myplace.now> wrote in part: >> The result from a study, be it double or triple blind, controlled or >> otherwise, should be expressed in the format of 'probabilities, degree [quoted text clipped - 16 lines] >carelessness and/or stupidity if the bone was broken in an accident while >driving drunk. Good observation, Bill. I'll add that randomized controlled trials do express their results in probabilities, contrary to fyfpoon's impression. At a minimum they report the comparison measures between the treatment and control groups, but they also report the statistical significance level the difference represents. The significance level is simply the probability that--if the active treatment does not differ from the control treatment--the observed results or more extreme results would occur by chance. This is expressed as the alpha level. Interestingly, acupuncture for tinnitus has been tested in many ways over decades. The differences between placebo and active treatment groups have been near zero and the significance level of the results tends to be around 1.0. There was a review of acupuncture for tinnitus treatment in a recent Cochrane review that covers the studies, BTW.  SignatureJim Chinnis / Warrenton, Virginia, USA Want to discuss Meniere's? See http://groups.yahoo.com/group/MenieresDG snipped... > Interestingly, acupuncture for tinnitus has been tested in many > ways over decades. Was the World Health Organization aware of these studies when they indicated that tinnitus was one of the 50 ills that could seek relief from acupuncture? FP The differences between placebo and active > treatment groups have been near zero and the significance level of > the results tends to be around 1.0. [quoted text clipped - 4 lines] > Jim Chinnis / Warrenton, Virginia, USA > Want to discuss Meniere's? See http://groups.yahoo.com/group/MenieresDG <snip> > Was the World Health Organization aware of these studies when they > indicated that tinnitus was one of the 50 ills that could seek relief > from acupuncture? > > FP People with any of these 50 ills could seek relief through prayer too, Francis. In fact, a comprehensive double blind study could compare the effectiveness of prayer to various recognized deities with prayer to an unrecognized, placebo deity. Be certain that a percentage could pray to a stone and report relief. I'm certain someone in the WHO knows this. > <snip> > [quoted text clipped - 9 lines] > unrecognized, placebo deity. Be certain that a percentage could pray to a > stone and report relief. I'm certain someone in the WHO knows this. So you think the people in WHO are less knowledgeable than you are. Do you? And that they just talk, talk...and talk. "fyfpoon@hotmail.com" <fyfpoon@hotmail.com> wrote in part: >snipped... >> Interestingly, acupuncture for tinnitus has been tested in many [quoted text clipped - 3 lines] >indicated that tinnitus was one of the 50 ills that could seek relief >from acupuncture? What probabilities did they report? SignatureJim Chinnis / Warrenton, Virginia, USA Want to discuss Meniere's? See http://groups.yahoo.com/group/MenieresDG > "fyfpoon@hotmail.com" <fyfpoon@hotmail.com> wrote in part: > [quoted text clipped - 7 lines] > > What probabilities did they report? I did not read that specific report but just quoted from this link http://www.internethealthlibrary.com/Health-problems/Tinnitus%20-%20researchAltT herapies.htm. After all, I am reading it as a layman. Regardless of the probabilities or the % of people with tinnitus that get relief from acupuncture, there are at least some T patients that do indicate their positive reaction to acupuncture. Otherwise the scientists and doctors at WHO would not have reported it. They don't just go by the 'here say', I suppose. On the other hand, I did talk to 4 others who have used acupuncture for their t in China. 2 spoke of this treatment highly, one so so and one negatively in the sense that there was no feedback on his part. I as a patient speak highly and very highly of it for *my* type of t. So how do i rationalize how I feel with the results from all that double-blind studies you have cited? Should I introduce my experience to other T patients? Or should I tell them that it is but placebo effect or just imaginary? Out of the sample that I used, 3 out of 5(including myself), is there any 'statistical truth' involved there? Have you had a survey to show the increasing (or decreasing) demand for acupuncture treatment from tinnitus patient in the US? These results could also shed light. The double-blind studies you have quoted are fine but are not necessarily the benchmark for 'truth'. Truth in my view is indisputable while the result from one study or one kind of studies could be contradicted by the result from another or another kind of study. In the strive for theoretical brilliance, many research scientists do tend to abuse the use of quantitative tools! That is something that has to be watched out carefully. No? FP > -- > Jim Chinnis / Warrenton, Virginia, USA > Want to discuss Meniere's? See http://groups.yahoo.com/group/MenieresDG "fyfpoon@hotmail.com" <fyfpoon@hotmail.com> wrote in part: >> "fyfpoon@hotmail.com" <fyfpoon@hotmail.com> wrote in part: >> [quoted text clipped - 38 lines] >scientists do tend to abuse the use of quantitative tools! That is >something that has to be watched out carefully. No? The double-blind studies are generally regarded as the benchmark of truth. SignatureJim Chinnis / Warrenton, Virginia, USA Want to discuss Meniere's? See http://groups.yahoo.com/group/MenieresDG Generally regarded? By who? What would happen if two studies of this kind conflicted each other? Has it ever happened? Truth to me may not be truth to you. Truth to me in relation to acupuncture for tinnitus is that it works for me and the UN-truth is what has come off the 'double-blind' studies you have cited around. What is truth? Or does truth exist at all? FP > Generally regarded? By who? What would happen if two studies of this > kind conflicted each other? Has it ever happened? [quoted text clipped - 5 lines] > > FP A conflict between two studies formulated like the one you posted would be as meaningless as either individual study. Everyone is entitled to their own opinion, Francis, but there is only one set of facts. You are confusing "opinion" with "truth" and your opinion is inconsistent with the definition of a controlled study. > Generally regarded? By who? What would happen if two studies of this > kind conflicted each other? Has it ever happened? [quoted text clipped - 5 lines] > > FP If you try acupuncture and feel relieved by it, then that's your truth, one which no one has a right to argue. If you present evidence that you claim is a "controlled study" and that turns out to be untrue, then that is open for disagreement. Susan Actually I don't think any 'controlled' study is valid at all, unless you make the assumption that every human body machine is more or less identical like auto parts from an assembly line. That is the biggest downfall of a controlled study. You do notice, however, that Jim has not replied as to the limitations of a double-blind study. FP > This isn't applicable to the results of any double blind study I've > reviewed. Every one I've seen publishes percentages of people in each [quoted text clipped - 4 lines] > carelessness and/or stupidity if the bone was broken in an accident while > driving drunk. It's applicable to every study I've ever read finding a causative role of fat in health and nutrition. Ignores the role of other confounding variables and completely gets the conclusions wrong. Susan Susan <Susan@nothankyou.com> wrote in part: >x-no-archive: yes > [quoted text clipped - 10 lines] >of fat in health and nutrition. Ignores the role of other confounding >variables and completely gets the conclusions wrong. That's why one needs to read the report and the statistics. Your very valid complaint doesn't usually change the fact that the randomized controlled diet studies do report the design, the outcome measures, and the PROBABILITIES (significance levels). The trouble is that the authors then go on to apply their biases in seeing what they expect to see and they then add a bunch of words that aren't justified. But the description, data, and probabilities are usually there. SignatureJim Chinnis / Warrenton, Virginia, USA Want to discuss Meniere's? See http://groups.yahoo.com/group/MenieresDG > Susan <Susan@nothankyou.com> wrote in part: > [quoted text clipped - 22 lines] > > But the description, data, and probabilities are usually there. Well, yeah, that's a point well taken. I learned just how bad the conclusions were just by looking even at the information at abstracts and seeing how it didn't support or didn't control enough to support the conclusions. OTOH, our researchers are not, IMO, reporting truthfully about their methods, either, in cases where they have a financial stake in the outcome, as with pharmaceuticals. In a vaccine study I know of, the investigators refused to accept reports of adverse reactions from study participants. The vaccine in question, LymeRix, was pulled off the market not long after being introduced with a splash. Lawsuits, lots of lawsuits by injured patients. Susan Susan <Susan@nothankyou.com> wrote in part: >> That's why one needs to read the report and the statistics. Your >> very valid complaint doesn't usually change the fact that the [quoted text clipped - 19 lines] >question, LymeRix, was pulled off the market not long after being >introduced with a splash. Lawsuits, lots of lawsuits by injured patients. There are, of course, liars everywhere and on all sides of most questions. Smaller studies with investigators who have conflicts of interest probabably should be considered anecdotal reports rather than clinical trials. Some larger clinical trials involve multiple investigators around the world. Many also arrive at their hypotheses and methods and publish them before the trial begins. There are lots of things wrong with the reporting of randomized trials, some of which are being addressed. But the failure to report probabilities--which was the claim that began this thread--is totally false. SignatureJim Chinnis / Warrenton, Virginia, USA Want to discuss Meniere's? See http://groups.yahoo.com/group/MenieresDG I recall that was how some dr.reported in this group.
|
Reply to this Message |
 Sign In
 Join
 My Latest Posts My Monitored Threads My Blog My Photo Gallery My Profile My Homepage Start New Thread Enable EMail Alerts Rate this Thread
|
©2008 Advenet LLC Privacy Policy - Terms of Use
This website includes both content owned or controlled by Advenet as well as content owned or controlled by third parties.