With H. Gilbert Welch's name on the cover, a book would have to go a long ways out of its way to make me unhappy.
This book made me very happy.
There are a lot of people and organization out there trying to get us to know our numbers and to take action based on those numbers. In much the same way that manufacturers of infant formula, doctors who gave advice regarding infant formula and so on said (and say) that "breast is best" and then blithely proceed to act as if breastfeeding is so unlikely and inconvenient an activity that no one will _actually_ engage in it (or, if they do, it won't work, or they'll stop soon, or they'll need to supplement or blah, blah, bleeping blah), advice about our numbers almost invariably says, stop smoking get exercise and eat healthfully -- and then proceeds to talk about medication options. And that's assuming they aren't going full bore on something even more invasive.
Woloshin et al provide a thin, readable and randomized control trial tested and published in a peer reviewed journal proven effective book (really! see page 105) on how to make sense of health research and advice from all sources. They show how to figure out if you've been told the group of people the research and/or advice applies to, what the starting risk is, what the modified risk is -- and how to recast that calculated per 1000 folk over ten years so have a consistent baseline. It's a beautiful, simple, clear and effective approach.
Buy a copy! Buy two and give one to a friend! Memorize this book, or at least consult the charts at the end frequently (which you can also download on the web -- I know I did that before I got the book). I don't think there's anyone alive and reading that would not benefit from this book.