I don't think that what we do is particularly advanced or special, we basically parse down situations into actors/attributes/tx principles, and then talk about whether it seems appropriate and what we can improve. We use it for our Data Review Board (a lightweight IRB).