Risk estimates in clinical research

Darshak Sanghavi of Slate has written a great comparison and contrast of relative risk, absolute risk, and number-needed-to-treat, and how the various estimates, though reflecting the same primary results, may lead to wildly differing perceptions of how "good" a certain treatment is--"Treat Me?The crucial health stat you've never heard of"

Bandolier also has a number of examples to help illustrate the utility of NNT estimates.

Thanks to Rachel for sending this article!