regex word boundaries in stata

#1
I would like to use \b to denote word boundaries but I'm not sure if it works in stata, although it does work on other platforms. Does anyone know how to get it to work in stata or failing that, how to use regex terms that do work in stata to simulate the effects of \b?

I think this guy has basically the same question but it seems like it was never answered: www.stata.com/statalist/archive/2006-12/msg00761.html