OK. I wrote so much last time that I was only going to say "thanks" to Tom's post. You put a face20on the trial participant - and that's the story we have to tell. That is why we have "human" trials - people who must cope with diseases like Parkinson's - who live its emotional ups and downs every day - otherwise we could just stop the trials at the primate (monkey) level.

There's a wonderful opportunity for us to ask an expert all of these questions first-hand on an upcoming webinar (the first I've heard of ever!) see announcement below:

Participating in Parkinson's Clinical Research: The Key to Becoming an Informed Study Volunteer

Wednesday, January 14, 2009, at 3:00 PM Eastern time

Featuring Kenneth Getz, M.B.A., author of The Gift of Participation: A Guide to Making Informed Decisions About Volunteering for a Clinical Trial

Click here to register:
http://event.netbriefings.com/event/.../register.html

I think this idea of testing of subtypes of Parkinson’s is brilliant. Greg Wasson came up with theidea (I believe) andwrote about it in an OpEd on the Pipeline site:
http://pdpipeline.org/advocacy/oped_..._july23008.htm

I'm not a statistician, and only took a couple of statistics courses at the Master's degree level, so if I am way off course, please correct me. We have small numbers of participants in phase I trials, because they are conducted to evaluate the safety of a treatment in humans. Any improvement of participants is a plus. Any unsafe results, and the trial is halted immediately.

The Phase II level is (or should be) carefully planned, with a requirement of "testing" enough participants to show that any improvement from the treatment didn't just happen per chance. Endpoints, or outcomes, must be hypothesized and projected. The IRB (Institutional Review Board) at each research institution decides what those expected results will be. There are statistical tests that must be done after all the data is gathered and a significance level at a minimum of 0.5 must be shown for the trial to be valid. Validity means the study tests what it said it was to test. Reliability is achieved when the study can be replicated with relatively the same results when repeated.

At present there is no test to “prove” that one has PD (with Multiple sclerosis, for instance, an MRI shows demyelinating or white patches visible on the brain). With Parkinson’s, we go by symptoms and the ruling out of other disorders.

So , I pose this question:
Is it ethical to “fake” the surgery a trial? Since PD is diagnosed via observation of symptoms, why doesn’t observing whether or not there is improvement of symptoms count?

I have been searching through some old books and online about this subtypes testing. There’s some really good stuff at this site: http://www.improvingmedicalstatistics.com/

If a study did test a treatment dividing them into subtypes of PD (e.g. tremor dominant vs rigidity or balance issues dyskinesia vs no dyskinesia), we would have to be very, very careful with how those groups were divided because itsays "the validity tends to be inversely proportional to the number of subgroups which are analyzed":

Inappropriate subgroup analysis can lead to ludicrous results.

Subgroup analysis, at times, can lead to findings that are incorrect. If a subgroup analysis results in an unexpected finding in outcome that is different from a highly significant and beneficial effect for the group as a whole, the subgroup analysis is often incorrect. In fact,it is more likely that the unexpected subgroup finding that runs counter to the group finding is simply not valid.
Source: Improving Medical Statistics

Read this very easy-to-comprehend article for a better explanation.:
http://www.improvingmedicalstatistic...s%20%20%20.pdf

This same article says, “A subgroup analysis which results in variance from the overall group outcome is more likely to be true if it involves a large subgroup and there are a very limited number of pre-specified analyses.” This means the Phase II recruiting of volunteers would be even greater in number. Recruitment would be even more difficult. It isn’t like people are knocking down doors to enter a trial where half of them will have sham surgery. But the results using subtype analysis would result in some real meat to chew on – a way of treating PD that’s never been done before. I sincerely believe this could turn the treatment of PD around.