Yet another take on PD types.

http://www.plosone.org/article/info%...l.pone.0070244

labrat,

Thanks for reporting this. I think it is important: one of the reasons that clinical trials fail is that the subjects don't all have the same illness. There may be a therapy out there which leads to a 10% improvement for a type "A" person, but a 10% worsening for a type "B". It probably wouldn't pass its clinical trials but, if you could prescribe it to the right group only, it would be useful.

Another advantage of cluster analysis is that it can help find the cause of the disease. It may, for instance, turn out that one of the clusters represents a genetic sub-group or a group affected by a common environmental toxin.

John