Give me the article and I will have a look. I am a believe in SCS in the non-work comp realm. But, my experience is that it is a palliative endeavor, not rehabilitative. By that I mean that doesn't improve return to work, functional status (SF-36, or ODI), and meta-analyses have shown - last year's NANS - that it doesn't diminish opioid use.
Here is the article (too big to attach PDF)
http://thejns.org/doi/full/10.3171/JNS/2008/108/2/0292
You are right. At second look at the article, patients didn't have functional improvement when compared to control, however, at five years they did have improved GPE, improved pain scores, and essentially a high satisfaction rate.
At 2 years, cost savings are the same as conventional therapy (the break even point) and at 3 years, there are cost savings.
A few interesting points about the article however. First of all, I find it interesting that 2 patients in the control side had complete resolution of symptoms (and each accounted for two so that is like 4 of them getting better spontaneuously.) I don't know about you, but I think that is very strange and outlier-ish. Second, the authors chose to use the post SCS trial scores for analysis rather than pre trial scores...that is another odity but it also scews the data away from significance - which by the way was at a P value of 0.06. The authors make this silly statement "and is no longer significant at 3 years of follow-up" - regarding pain.
Are you serious? So what you are telling me mr author, is that the arbitrary number you magically picked of 0.05 IS significant, but since you picked 0.05, that means 0.06% isn't significant?
Researchers.....blah. Stupid stupid researchers.
I agree it is a small study, but it is better data than mental health and physical therapy and OT with CRPS - and we insist on that right? (I know, I know....graded motor imagery has class I evidence...but who really does that? I don't think I've ever even met a physical therapist that knew what that was.)