Hi Rachel, This is a fascinating take of a problem that has been vexing me for year. Indeed, I've been working on a technology solution to 'guarantee' that the the data _will_ circle back. We've been developing a curation software that actually analyzes all the documents that you normally would have had the opportunity to read and employs smart natural-language-processing to refine them into a manageable list. I'll send you a note when gnowit.com is ready to step out of stealth mode, and would love some feedback on whether you found this useful. Best, Shahzad