evidence based practice or practice based evidence?

I did not finish yesterday's rant so here is the next installment. In the car today I caught part of a really interesting discussion on Radio 4 which was about obesity research. The academic was saying how that within obesity research there is a trend being led by Government to drop high quality search in favour in favour of what was called "practice based evidence" because producing high quality research was too difficult. The academic said that practice based evidence is not evidence, yet it rang loud and clear with me because so much of what passes off within physiotherapy as "clinical effectiveness" is the same stuff, so-called evidence of effect produced from everyday practice. One of the big flaws which has crept into the profession is the use of sloppy outcome measures, adopted because they have the magic word "validated" attached to them yet often inappropriate to what needs to be measured or able to show change but unable to demonstrate this was as a result of physiotherapy (why is there always the assumption that all we need to do is to demonstrate change in the right direction on an outcome measure?) I have big problems with MYMOP, an increasingly popular measure with physios, popoular because it can be used in a wide range of circumstances but something which fails to make common sense. It's a bit like measuring patient satisfaction, the measure of which is usually high for most (therefore by assumption are we all doing a good job even whan there is evidence to the contrary?) I also have reservations because an outcome measure can achieve validation in statistical terms but does that then mean it is always going to be useful within clinical situations very different from those in which it was developed? Am I talking about staining the limits of validity?

As usual I've got more questions than answers but I thought that's part of the reason for having a blog.